<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Saif Ali Shaik on Medium]]></title>
        <description><![CDATA[Stories by Saif Ali Shaik on Medium]]></description>
        <link>https://medium.com/@saifshines?source=rss-6eeb07f34fec------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*bn4R7kARL4R4JtAIGCPaJA.jpeg</url>
            <title>Stories by Saif Ali Shaik on Medium</title>
            <link>https://medium.com/@saifshines?source=rss-6eeb07f34fec------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 12:26:56 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@saifshines/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Your B2B billing system is asking the wrong question]]></title>
            <link>https://saifshines.medium.com/b2b-billing-system-design-architecture-397c569986e0?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/397c569986e0</guid>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[payments]]></category>
            <category><![CDATA[system-design-concepts]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[authentication]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Thu, 06 Aug 2026 14:19:50 GMT</pubDate>
            <atom:updated>2026-08-06T14:19:50.831Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>Your billing system wants to know who’s paying. It should be asking which company is.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZFF7f_KXnpTg0VIJrKGLyA.png" /><figcaption>Chargebee X Scalekit</figcaption></figure><p>Every B2B SaaS product eventually has the same argument with itself.</p><ul><li>Your <strong>authentication</strong> system knows about <strong>users</strong>.</li><li>Your <strong>billing provider</strong> knows about <strong>customers</strong>.</li><li>Your <strong>product</strong> is sold to <strong>companies</strong>.</li></ul><p>These are three different nouns, and there is exactly one place in your architecture where they have to be reconciled: your application code, usually written at speed, usually by whoever picked up the ticket.</p><p>For a while it doesn’t matter. Early accounts have one person in them, so user and company are the same thing and the mapping is trivial.</p><p>The trouble arrives with your first real customer: five seats, a procurement process, an admin who isn’t the person who signed up. Suddenly the identity you keyed billing to is a junior engineer who joined a Slack invite link, and the invoice is going to someone who left in March.</p><blockquote>That’s not a billing bug. It’s a modelling decision made months earlier, coming due.</blockquote><p>I want to walk through what it looks like to <strong>make that decision correctly</strong>, using a reference application we built at <a href="https://www.scalekit.com/">Scalekit</a>: a <strong>Next.js 14 app</strong> that wires Scalekit’s<strong> </strong><a href="https://www.scalekit.com/user-management"><strong>organization-led authentication</strong></a> to Chargebee’s<strong> subscription billing</strong>. <a href="https://github.com/scalekit-developers/saas-auth-chargebee-example">The code is on GitHub</a>, but the interesting part is the constraint it enforces.</p><h3>One commitment, made once</h3><p>Here is the entire design, in a sentence:</p><blockquote>The<strong> organization ID </strong>from a <strong>validated access token </strong>is the <strong>billing reference.</strong></blockquote><p>Not just a <strong>user ID</strong>. Not a random <strong>customer ID</strong> passed up from the frontend. Not a tenant slug in the URL. The <strong>oid</strong> claim, read from a token the server has independently validated, on every single billing request.</p><p>That sounds almost too small to be an architecture. It isn’t. Every other decision in the system is a consequence of it, and the consequences are what make it worth writing down.</p><p>The database schema shows this most clearly:</p><pre>export const organization = sqliteTable(&#39;organization&#39;, {<br>  id: text(&#39;id&#39;).primaryKey(),<br>  chargebeeCustomerId: text(&#39;chargebee_customer_id&#39;).unique(),<br>  displayName: text(&#39;display_name&#39;),<br>  updatedAt: integer(&#39;updated_at&#39;, { mode: &#39;timestamp&#39; }),<br>});</pre><p>Look at the primary key. It isn’t an auto-increment integer with a Scalekit ID stored alongside it. organization.id <em>is</em> the Scalekit organization ID. There&#39;s no translation layer, because a translation layer is a place where the two systems can disagree.</p><p>The subscription table carries the same value in a <strong>referenceId</strong> column, and every route that touches billing, create, list, portal, cancel, resolves through it.</p><h3>The guard clause that carries the system</h3><p>Login is unremarkable <strong>OAuth</strong>: generate state, store it for <strong>CSRF protection</strong>, redirect to <strong>Scalekit</strong>, validate state on the way back, exchange the code, drop an <strong>HttpOnly</strong> session cookie.</p><p>The part worth reading twice is <em>what happens on every billing request afterwards</em>.</p><pre>const claims = await client.validateToken(accessToken);<br>const organizationId = claims.oid;<br>if (!organizationId) {<br>  throw new SessionError(403, &#39;Organization context required for billing&#39;);<br>}</pre><p>Two things are happening here, and only one of them is obvious.</p><p>The obvious one:<strong> no organization context, no billing.</strong></p><p>A personal account, a token that predates organization support, a user who somehow hasn’t been assigned — all rejected before anything touches Chargebee.</p><p>The subtle one: <strong>the route calls </strong><strong>validateToken() rather than reading the cookie&#39;s contents.</strong> It doesn&#39;t trust its own session cookie as a source of claims. It re-validates.</p><p>And in the reference app, the middleware guarding /billing and /dashboard only checks whether the cookie <em>exists</em>. That&#39;s deliberate, and I think it&#39;s the most transferable idea in the whole project. <strong>Middleware that checks for the presence of a session is doing routing, not authorization.</strong> It sends unauthenticated users to the login page. It does not decide what an authenticated user may do. That decision belongs in the handler, next to the action, where the specific resource is in scope.</p><p>Plenty of applications collapse those two jobs into one, and the collapse is invisible until the day someone finds a route the middleware pattern didn’t match.</p><h3>Where cross-tenant bugs actually come from</h3><p>There’s a route in the app that accepts a referenceId in the request body. On first read that looks like the vulnerability I&#39;ve spent several paragraphs warning about. It isn&#39;t, and the reason is instructive:</p><pre>const referenceId = body.referenceId ?? ctx.organizationId;<br>const authorized = await authorizeReference({<br>  userId: ctx.userId,<br>  organizationId: ctx.organizationId,<br>  referenceId,<br>  action: &#39;create&#39;,<br>});<br><br>if (!authorized) {<br>  return NextResponse.json({ error: &#39;Forbidden&#39; }, { status: 403 });<br>}</pre><p>The client-supplied value is <em>defaulted</em> from the verified session and then <em>explicitly authorized against it</em>. Accepting input from a client is fine. Acting on it without checking is not. The difference between a billing endpoint and an insecure direct object reference is precisely this function call.</p><p>It’s also the hook where real products put their role logic. A member of an organization probably shouldn’t be able to cancel the company’s subscription. authorizeReference is where you say so — once, in one place, rather than scattered across six route handlers.</p><h3>The race condition you will otherwise meet in production</h3><p>Hosted checkout is asynchronous in a way that’s easy to underestimate.</p><p>The user completes payment. <strong>Chargebee</strong> redirects them back to your app. Chargebee also sends a webhook. These are two independent events on two independent schedules, and the ordering is not guaranteed. Sometimes the browser redirect wins — the user is looking at your billing page before your server has any idea a subscription exists.</p><p>The naive version of this app renders an empty billing page to a customer who just paid. They email support. Support asks engineering. Engineering says “it works on my machine,” because on a fast local tunnel the webhook usually wins.</p><p>The fix in the reference app is small and slightly counterintuitive: create a local subscription row <em>before</em> opening checkout, in a future state.</p><pre>localSub = await createFutureSubscription({<br>  referenceId,<br>  chargebeeCustomerId: customerId,<br>});</pre><p>Now your app has an identifier for this subscription before Chargebee does. Whichever event arrives first, there’s a row waiting to be updated. The race doesn’t get won or lost; it stops being a race.</p><p>The success redirect attempts a sync of its own, and if the webhook already handled it, that sync is a harmless no-op. Both orderings converge on the same state. That’s the property you want from anything asynchronous.</p><h3>Webhooks, and the discipline of resolution</h3><p>Chargebee webhooks are the source of truth for subscription state in this design. Your local tables are a cache, and the app never pretends otherwise.</p><p>Before parsing anything, the Scalekit webhook route reads the raw request body and verifies the signature against it:</p><pre>const rawBody = await req.text();<br>const isValid = client.verifyWebhookPayload(<br>  secret,<br>  headersToRecord(req),<br>  rawBody<br>);</pre><p>Raw body first, then verify, then dispatch asynchronously and return 200 quickly. Parsing before verifying means you’ve processed attacker-controlled JSON, and re-serializing a parsed body for signature checking is a well-known way to introduce mismatches.</p><p>On the Chargebee side, the app handles the usual lifecycle set: created, activated, started, changed, renewed, cancellation-removed, cancelled, plus customer deletion — and resolves each event to a local row through four ordered fallbacks:</p><ol><li>An existing Chargebee subscription ID already on the local row</li><li>subscription.meta_data.subscriptionId</li><li>customer.meta_data.pendingSubscriptionId</li><li>The future row for that organization reference</li></ol><p>Four paths looks like paranoia. It reads differently once you’ve received a webhook for a subscription your application has genuinely never heard of: a retry after a deploy, an event created in the Chargebee dashboard by someone on the finance team, a replay from a queue. The metadata written at customer-creation time is what makes fallbacks two and three possible, which is why the organization ID gets stamped into Chargebee from the very first API call:</p><pre>const { customer } = await chargebee.customer.create({<br>  company: displayName ?? undefined,<br>  email: email ?? undefined,<br>  preferred_currency_code: &#39;USD&#39;,<br>  meta_data: {<br>    organizationId,<br>    customerType: &#39;organization&#39;,<br>  },<br>});</pre><p>Once resolved, Chargebee’s state is copied into local columns- status, period boundaries, trial dates, cancellation timestamp, seat count, and the local subscription items are replaced wholesale rather than diffed. Replacing is boring and correct. Diffing subscription items is how you end up with a phantom add-on that nobody can explain.</p><p>The payoff is that /billing never calls Chargebee to render. It reads local state through two endpoints. No third-party latency in your page load, no rate limits in your critical path, no outage in someone else&#39;s system taking down your billing screen.</p><h3>The seam where your product goes</h3><p>Everything application-specific lives in one file, lib/subscription-hooks.ts, stubbed out with console logs:</p><p>onCustomerCreate, onSubscriptionCreated, onSubscriptionComplete, onSubscriptionUpdated, onSubscriptionDeleted, onSubscriptionCancel, onTrialStart, onTrialEnd, onAuthorizeReference.</p><p>Grant entitlements in onSubscriptionComplete. Fire product analytics in onTrialStart. Trigger a customer success workflow in onTrialEnd. Enforce admin-only billing in onAuthorizeReference.</p><p>Isolating these matters more than it might seem. Entitlement logic has a strong tendency to metastasize- a feature check here, a seat count there, a plan comparison in a component somewhere, until nobody can answer “what does this customer actually have access to?” without reading four files. One module with named hooks is a weak defence against that, but it’s a defence.</p><h3>What this is not</h3><p>I’d rather be direct about the limits than have someone deploy a demo.</p><p>It’s a reference application, deliberately small so the boundaries stay legible. It uses SQLite. It uses Drizzle’s db:push rather than migration files. The hooks are stubs. There are no tests. It includes developer-facing session inspection pages that expose raw tokens. And the Chargebee webhook route logs a warning and skips Basic Auth entirely when the credentials aren&#39;t configured, which is convenient for local development and unacceptable anywhere else.</p><p>Before this pattern goes to production: configure both webhook endpoints’ authentication, keep the Scalekit webhook secret mandatory, write real migrations, add tests covering authorization failures and webhook replay specifically, and remove the debug surfaces.</p><p>The value on offer isn’t a billing system. It’s the shape of one, with every seam labelled.</p><h3>Why the division of labour holds</h3><p>Scalekit validates identity and organization membership. Chargebee runs checkout, subscription lifecycle, and the customer portal. Your application owns an organization reference and a small local model- three tables and a handful of route handlers.</p><p>That’s a defensible split, not because building auth or billing is impossible, but because neither is your product, and both are the kind of system that quietly demands maintenance forever. Tax logic changes. SSO protocols have edge cases. Dunning is its own discipline.</p><p>What your team should own is the reconciliation between them, because that part is genuinely specific to how you sell.</p><p>And it holds one invariant in place, which is the thing I’d want a reader to leave with:</p><h4><strong>Every billing action begins from a verified organization, never from a client-supplied customer or subscription ID.</strong></h4><p>Decide that in week two and it costs you a guard clause. Decide it in year two and it costs you a migration, a security review, and an awkward conversation about how long the gap was open.</p><blockquote><strong>Full technical walkthrough:</strong> <a href="https://www.scalekit.com/blog/scalekit-chargebee-organisation-billing">scalekit.com/blog/scalekit-chargebee-organisation-billing</a></blockquote><blockquote><strong>Source code:</strong> <a href="https://github.com/scalekit-developers/saas-auth-chargebee-example">github.com/scalekit-developers/saas-auth-chargebee-example</a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=397c569986e0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Add Enterprise SSO to Your Next.js App in Minutes Using Claude Code & Scalekit]]></title>
            <link>https://javascript.plainenglish.io/add-enterprise-sso-to-your-next-js-app-in-minutes-using-claude-code-scalekit-f7d2e01c1189?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/f7d2e01c1189</guid>
            <category><![CDATA[claude-code]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[nextjs]]></category>
            <category><![CDATA[typescript]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Tue, 24 Mar 2026 07:04:23 GMT</pubDate>
            <atom:updated>2026-03-24T07:04:23.772Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iRcpZoMT5vDV7KIYhSaLhQ.png" /><figcaption>Add SSO to NextJS using Scalekit’s Marketplace Plugins for Claude Code</figcaption></figure><p><strong>The old way</strong>: Spend hours reading OAuth documentation, configuring SAML endpoints, building login flows, handling token validation, and creating admin portals for your enterprise customers.</p><p><strong>The new way</strong>: Describe what you want in plain English. Let Claude Code and Scalekit’s plugin do the rest.</p><p>This guide shows you how to add production-ready enterprise Single Sign-On to a Next.js application using Claude Code’s plugin ecosystem — going from zero to a working SSO flow with customer self-service portal in under 10 minutes.</p><h3>What You’re Building</h3><p>By the end of this guide, you’ll have:</p><p><strong>1. Enterprise SSO Authentication Flow</strong></p><ul><li>Users enter their work email (user@company.com)</li><li>Automatic redirect to their organization’s identity provider (Microsoft Entra ID, Okta, Google Workspace, etc.)</li><li>Secure callback handling with token validation</li><li>Session creation with verified user identity</li></ul><p><strong>2. Self-Service Admin Portal</strong></p><ul><li>Your B2B customers configure their own IDP connections</li><li>No engineering involvement needed for enterprise onboarding</li><li>Support for SAML and OIDC protocols</li><li>Real-time connection status monitoring</li></ul><p><strong>3. Production-Ready Code</strong></p><ul><li>Environment-based configuration</li><li>Proper error handling</li><li>Security best practices baked in</li><li>Next.js App Router patterns</li></ul><h3>Understanding the Plugin Architecture</h3><p>Before we dive in, let’s understand what makes this possible.</p><h3>What Are Claude Code Plugins?</h3><p>Claude Code plugins are <strong>packaged bundles of capabilities</strong> that extend what Claude can do. Think of them as specialized skill packs that teach Claude how to work with specific services, frameworks, or patterns.</p><p>A plugin can contain:</p><ul><li><strong>Skills</strong>: Context-aware instruction sets Claude automatically uses when relevant</li><li><strong>MCP Servers</strong>: Connections to external APIs and services via Model Context Protocol</li><li><strong>Commands</strong>: Slash commands you invoke manually (e.g., /deploy)</li><li><strong>Subagents</strong>: Specialized AI agents for specific tasks (e.g., security review)</li><li><strong>Hooks</strong>: Scripts that run on specific events (pre-commit, post-deployment)</li></ul><h3>How Scalekit’s Plugin Works</h3><p>The Scalekit Auth Stack plugin gives Claude Code:</p><ol><li><strong>Documentation Access</strong>: Real-time access to Scalekit’s implementation guides via MCP</li><li><strong>API Integration</strong>: Ability to query your Scalekit environment, retrieve config, manage connections</li><li><strong>Implementation Patterns</strong>: Battle-tested code patterns for SSO flows, session management, and security</li><li><strong>Framework Knowledge</strong>: How to properly integrate with Next.js, Express, FastAPI, and other frameworks</li></ol><p>When you ask Claude to “add SSO,” it:</p><ul><li>Reads your existing codebase structure</li><li>Pulls relevant Scalekit documentation</li><li>Queries your Scalekit environment via MCP</li><li>Generates complete, working code that follows your project’s patterns</li></ul><h3>Prerequisites</h3><p><strong>Required:</strong></p><ul><li>A Next.js application (even a fresh create-next-app works)</li><li><a href="https://code.claude.com/docs/en/quickstart">Claude Code installed</a> and configured</li><li>A <a href="https://scalekit.com/">Scalekit account</a> with dashboard access</li></ul><p><strong>You’ll need from Scalekit:</strong></p><ul><li>Environment URL (e.g., <a href="https://yourenv.scalekit.com)/">https://yourenv.scalekit.com)</a></li><li>Client ID (from your Scalekit dashboard)</li><li>Client Secret (generate in dashboard, keep secure)</li></ul><p><strong>Time investment:</strong> ~10 minutes</p><h3>Step-by-Step Implementation</h3><h4>Step 1: Install the Scalekit Plugin</h4><p>Open your terminal and start the Claude Code REPL:</p><pre>claude</pre><p>Add the Scalekit Auth Stack marketplace:</p><pre>/plugin marketplace add scalekit-inc/claude-code-authstack</pre><p>You should see confirmation:</p><pre>/plugin marketplace add scalekit-inc/claude-code-authstack</pre><pre>Successfully added marketplace: scalekit-auth-stack</pre><p>Now install the Modular SSO plugin:</p><pre>/plugin install modular-sso@scalekit-auth-stack</pre><p><strong>What just happened?</strong></p><ul><li>Claude Code now has access to Scalekit’s SSO implementation knowledge</li><li>The plugin includes MCP servers that can query your Scalekit environment</li><li>Skills are loaded that guide code generation for auth flows</li></ul><p><strong>Verify installation:</strong></p><pre>/plugins</pre><p>You should see modular-sso in your installed plugins list.</p><h4>Step 2: Navigate to Your Project</h4><p>Exit the REPL (Ctrl+D or exit) and navigate to your Next.js project:</p><pre>cd /path/to/your-nextjs-app</pre><p>Start Claude Code in your project:</p><pre>claude</pre><h4>Step 3: Give Claude the Objective</h4><p>Here’s where the magic happens. Paste this prompt:</p><pre>My goal is to add SSO to my application and also let my customers <br>configure SSO connections with their IDP using an admin portal.</pre><p><strong>Watch Claude work:</strong></p><ol><li><strong>Skill Detection</strong>: Claude automatically identifies the Modular SSO skill is relevant</li><li><strong>Codebase Analysis</strong>: Reads your package.json, route structure, and existing patterns</li><li><strong>Documentation Retrieval</strong>: Fetches Scalekit’s implementation guide via MCP</li><li><strong>Environment Discovery</strong>: Queries your Scalekit account to get environment details</li><li><strong>Code Generation</strong>: Creates all necessary files following Next.js conventions</li></ol><p>You’ll see output like:</p><pre>Loading skill: Modular SSO<br>Analyzing project structure...<br>Reading Next.js configuration...<br>Fetching Scalekit environment configuration...<br>Found environment: https://yourenv.scalekit.com<br>Retrieved Client ID: sk_live_...</pre><h3>Step 4: Approve the Redirect URI</h3><p>Claude will ask permission to configure a <strong>redirect URI</strong> in your Scalekit environment:</p><pre>I need to add a redirect URI to your Scalekit environment:<br>http://localhost:3000/api/auth/callback</pre><pre>Allow? [y/n]</pre><p>Type y and press Enter.</p><p><strong>Why this matters:</strong><br>The redirect URI is where Scalekit sends users after they authenticate with their IDP. It must be registered for security — preventing malicious sites from receiving your auth callbacks.</p><h3>Step 5: Add Your Client Secret</h3><p>Claude will generate a .env.local file with:</p><pre>SCALEKIT_ENV_URL=https://yourenv.scalekit.com<br>SCALEKIT_CLIENT_ID=sk_live_abc123...<br>SCALEKIT_CLIENT_SECRET=</pre><p>Open your Scalekit dashboard, navigate to <strong>Settings → API Keys</strong>, and generate a new Client Secret.</p><p><strong>Copy it and add to </strong><strong>.env.local:</strong></p><pre>SCALEKIT_CLIENT_SECRET=sk_secret_xyz789...</pre><p><strong>Security note</strong>: Never commit this file. Claude should have added it to .gitignore automatically.</p><h3>Step 6: Review Generated Files</h3><p>Claude will have created:</p><pre>your-app/<br>├── .env.local                          # Environment configuration<br>├── app/<br>│   ├── api/<br>│   │   └── auth/<br>│   │       ├── [org]/<br>│   │       │   └── route.ts           # SSO initiation<br>│   │       └── callback/<br>│   │           └── route.ts           # OAuth callback handler<br>│   ├── login/<br>│   │   └── page.tsx                   # Login page with email input<br>│   └── sso-settings/<br>│       └── page.tsx                   # Admin portal for IDP config<br>├── lib/<br>│   └── scalekit.ts                    # Scalekit client initialization<br>└── middleware.ts                       # Session validation (optional)</pre><p><strong>Key file: </strong><strong>lib/scalekit.ts</strong></p><pre>import { ScalekitClient } from &#39;@scalekit-sdk/node&#39;;</pre><pre>export const scalekit = new ScalekitClient(<br>  process.env.SCALEKIT_ENV_URL!,<br>  process.env.SCALEKIT_CLIENT_ID!,<br>  process.env.SCALEKIT_CLIENT_SECRET!<br>);</pre><p><strong>Key file: </strong><strong>app/api/auth/callback/route.ts</strong></p><pre>import { scalekit } from &#39;@/lib/scalekit&#39;;<br>import { NextRequest } from &#39;next/server&#39;;</pre><pre>export async function GET(request: NextRequest) {<br>  const searchParams = request.nextUrl.searchParams;<br>  const code = searchParams.get(&#39;code&#39;);<br>  const error = searchParams.get(&#39;error&#39;);</pre><pre>  if (error) {<br>    return Response.redirect(&#39;/login?error=&#39; + error);<br>  }</pre><pre>  if (!code) {<br>    return Response.redirect(&#39;/login?error=no_code&#39;);<br>  }</pre><pre>  try {<br>    // Exchange code for user identity<br>    const result = await scalekit.authenticateWithCode(<br>      code,<br>      request.nextUrl.origin + &#39;/api/auth/callback&#39;<br>    );</pre><pre>    // Create session with user data<br>    // (Implementation depends on your session strategy)</pre><pre>    return Response.redirect(&#39;/dashboard&#39;);<br>  } catch (err) {<br>    console.error(&#39;Auth error:&#39;, err);<br>    return Response.redirect(&#39;/login?error=auth_failed&#39;);<br>  }<br>}</pre><h3>Step 7: Install Dependencies</h3><p>Claude should have updated package.json, but verify:</p><pre>pnpm install<br># or npm install, or yarn install</pre><h3>Testing with the IDP Simulator</h3><p>You don’t need a real enterprise IDP to test this. Scalekit provides a built-in simulator.</p><h3>Start Your Dev Server</h3><pre>pnpm dev</pre><p>Navigate to http://localhost:3000—you should be redirected to /login.</p><h3>Use the Test Organization</h3><p>In your Scalekit dashboard:</p><ol><li>Go to <strong>Organizations</strong></li><li>Find the <strong>Test Organization</strong> (created by default)</li><li>Note it’s configured with example.com as an organization domain</li><li>The IDP is set to <strong>IDP Simulator</strong> (simulates Okta/Microsoft/Google)</li></ol><h3>Test the Flow</h3><ol><li>On the login page, enter: yourname@example.com</li><li>Click <strong>Continue</strong></li><li>You’ll be redirected to the IDP Simulator (looks like a real enterprise login page)</li><li>Enter any name in the form</li><li>Click <strong>Simulate Login</strong></li></ol><p><strong>What happens:</strong></p><ul><li>Scalekit matches example.com to the Test Organization</li><li>Redirects to the configured IDP (Simulator)</li><li>Simulator “authenticates” the user</li><li>Redirects back to your app at /api/auth/callback?code=...</li><li>Your app exchanges the code for user identity</li><li>Session is created, user is logged in</li></ul><h3>Check the Admin Portal</h3><p>Navigate to <a href="http://localhost:3000/sso-settings">http://localhost:3000/sso-settings</a></p><p>You should see:</p><ul><li><strong>Connected Identity Providers</strong> section</li><li>Status of each IDP (Enabled/Disabled)</li><li>Available IDPs to configure (Microsoft Entra ID, Okta, Google, etc.)</li></ul><p>This is what your enterprise customers will use to self-serve their SSO configuration.</p><h3>How It Actually Works</h3><p>Let’s demystify what just happened.</p><h4>The SSO Flow</h4><pre>1. User visits app<br>   ↓<br>2. Redirected to /login<br>   ↓<br>3. User enters work email: user@acme.com<br>   ↓<br>4. App calls Scalekit: &quot;Get SSO URL for acme.com&quot;<br>   ↓<br>5. Scalekit checks: Is there an org with domain acme.com?<br>   ↓<br>6. If yes: Return IDP login URL for that org<br>   ↓<br>7. User redirected to their company&#39;s IDP (Okta/Microsoft/etc)<br>   ↓<br>8. User authenticates at IDP<br>   ↓<br>9. IDP redirects back to: yourapp.com/api/auth/callback?code=xyz<br>   ↓<br>10. App exchanges code with Scalekit for user identity<br>    ↓<br>11. Scalekit returns: { email, name, organization, ... }<br>    ↓<br>12. App creates session, user is logged in</pre><h4>Organization-Based Routing</h4><p>The key insight: <strong>Scalekit routes authentication based on email domain</strong>.</p><p>When a user enters jane@acme.com:</p><ul><li>Scalekit looks up which organization owns acme.com</li><li>Retrieves that organization’s configured IDP</li><li>Returns the appropriate login URL</li></ul><p>This is why your B2B customers can each use different IDPs — routing happens automatically per organization.</p><h4>The Admin Portal</h4><p>The self-service portal uses Scalekit’s management APIs to:</p><ol><li><strong>List Organizations</strong>: Show the customer their organization details</li><li><strong>Configure Connections</strong>: Let them set up SAML/OIDC with their IDP</li><li><strong>Test Connections</strong>: Validate the setup before going live</li><li><strong>Manage Domains</strong>: Add/remove email domains for their org</li></ol><p>All without your engineering team touching anything.</p><h4>Security Considerations Built In</h4><p>The generated code includes:</p><ul><li><strong>PKCE (Proof Key for Code Exchange)</strong>: Protects against authorization code interception</li><li><strong>State Parameter</strong>: Prevents CSRF attacks</li><li><strong>Token Validation</strong>: Ensures codes haven’t been tampered with</li><li><strong>Environment-Based Config</strong>: Secrets never in version control</li><li><strong>HTTPS Enforcement</strong>: Redirect URIs must use HTTPS in production</li></ul><h3>Production Considerations</h3><h4>Environment Variables</h4><p>For production, set these in your hosting platform (Vercel, AWS, etc.):</p><pre>SCALEKIT_ENV_URL=https://yourenv.scalekit.com<br>SCALEKIT_CLIENT_ID=sk_live_...<br>SCALEKIT_CLIENT_SECRET=sk_secret_...</pre><h4>Update Redirect URIs</h4><p>In your Scalekit dashboard, add production callback URLs:</p><pre>https://yourdomain.com/api/auth/callback</pre><h4>Session Management</h4><p>Claude generates a basic session setup. For production, consider:</p><ul><li><strong>Session Storage</strong>: Redis, database, or encrypted cookies</li><li><strong>Session Lifetime</strong>: How long until re-authentication required</li><li><strong>Token Refresh</strong>: For long-lived sessions</li><li><strong>Multi-Device</strong>: Handling concurrent sessions</li></ul><h4>Error Handling</h4><p>Add user-friendly error messages:</p><pre>// app/login/page.tsx<br>const errorMessages = {<br>  no_code: &#39;Authentication failed. Please try again.&#39;,<br>  auth_failed: &#39;Could not verify your identity.&#39;,<br>  invalid_email: &#39;Please use your work email address.&#39;,<br>  no_org: &#39;Your organization has not set up SSO yet.&#39;<br>};</pre><h3>Monitoring</h3><p>Track:</p><ul><li>Failed authentication attempts</li><li>Organization-wise SSO usage</li><li>IDP connection health</li><li>Session creation/destruction</li></ul><p>Scalekit provides webhooks for these events.</p><h3>Alternative: Using Other AI Coding Agents</h3><p>Scalekit’s Auth Stack works with 40+ AI coding agents, not just Claude Code.</p><h4>GitHub Copilot CLI</h4><pre># Install marketplace<br>copilot plugin marketplace add scalekit-inc/github-copilot-authstack</pre><pre># Install plugin<br>copilot plugin install modular-sso@scalekit-auth-stack</pre><pre># Generate implementation<br>copilot &quot;Add Scalekit SSO to my app with admin portal for customer IDP configuration&quot;</pre><h4>Cursor</h4><p><strong>Option</strong>: Use Claude Code marketplace (works now)</p><pre># In Claude Code REPL<br>/plugin marketplace add scalekit-inc/claude-code-authstack</pre><pre># Cursor automatically picks up Claude Code plugins<br># Open Cursor → Settings → Plugins → Enable Modular SSO</pre><p>Then use Cursor’s chat with Cmd+L / Ctrl+L:</p><pre>Add Scalekit SSO with customer admin portal for IDP configuration</pre><h4>Any Agent via Vercel Skills</h4><pre># Interactive install<br>npx skills add scalekit-inc/skills</pre><pre># Or list and pick<br>npx skills add scalekit-inc/skills --list</pre><pre># Install specific skill<br>npx skills add scalekit-inc/skills --skill modular-sso</pre><pre># Global install for all projects<br>npx skills add scalekit-inc/skills --all --global</pre><p>Supported agents: Windsurf, Cline, Gemini CLI, OpenCode, Codex, and 30+ others.</p><h3>Why This Matters</h3><p><strong>Traditional SSO implementation:</strong></p><ul><li>Read OAuth 2.0 / SAML spec (hours)</li><li>Study IDP-specific quirks (hours)</li><li>Write authorization URL generation (30 min)</li><li>Handle callback and token exchange (1 hour)</li><li>Build session management (2+ hours)</li><li>Create admin UI for IDP config (3+ hours)</li><li>Test with multiple IDPs (hours)</li><li>Debug edge cases (days)</li></ul><p><strong>Total: 1–2 weeks for an experienced developer</strong></p><p><strong>With Claude Code + Scalekit:</strong></p><ul><li>Install plugin (1 minute)</li><li>Describe what you want (30 seconds)</li><li>Review generated code (5 minutes)</li><li>Test with simulator (2 minutes)</li></ul><p><strong>Total: ~10 minutes</strong></p><p>The code quality is production-ready because:</p><ol><li>The plugin encodes battle-tested patterns from thousands of implementations</li><li>It follows your framework’s conventions automatically</li><li>Security best practices are baked in</li><li>Scalekit handles the complex parts (SAML parsing, token validation, IDP integrations)</li></ol><h3>Conclusion</h3><p>Enterprise SSO used to be a feature that delayed product launches. Between understanding OAuth flows, handling SAML complexity, and building customer onboarding portals, it was weeks of work.</p><p>AI coding agents with specialized plugins collapse that timeline to minutes. But this isn’t about AI writing code faster — it’s about <strong>encoding expert knowledge</strong> into reusable skills that handle entire feature implementations.</p><p>The Scalekit plugin doesn’t just generate boilerplate. It:</p><ul><li>Understands your codebase structure</li><li>Queries your live environment</li><li>Follows security best practices</li><li>Creates complete, working flows</li></ul><p>This is the new paradigm: describe the outcome, let specialized agents handle the implementation details.</p><p><strong>Next steps:</strong></p><ul><li><a href="https://docs.scalekit.com/">Scalekit Documentation</a></li><li><a href="https://code.claude.com/docs/en/plugins">Claude Code Plugin Development</a></li><li><a href="https://github.com/scalekit-inc/claude-code-authstack">Scalekit Auth Stack GitHub</a></li><li><a href="https://github.com/scalekit-inc/scalekit-nextjs-example">Example Implementation</a></li></ul><p><strong>Try it yourself</strong>: The fastest way to understand this isn’t reading — it’s doing. Spin up a fresh Next.js app, install the plugin, and watch Claude build your SSO flow in real-time.</p><p>The future of development isn’t writing less code. It’s describing what you want and having specialized agents that actually understand your domain build it for you.</p><p><em>Have questions? Join the </em><a href="https://scalekit-community.slack.com/"><em>Scalekit’s Slack</em></a><em> or the </em><a href="https://discord.gg/claude-code"><em>Claude Code community</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f7d2e01c1189" width="1" height="1" alt=""><hr><p><a href="https://javascript.plainenglish.io/add-enterprise-sso-to-your-next-js-app-in-minutes-using-claude-code-scalekit-f7d2e01c1189">Add Enterprise SSO to Your Next.js App in Minutes Using Claude Code &amp; Scalekit</a> was originally published in <a href="https://javascript.plainenglish.io/">JavaScript in Plain English</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Global Apps Framework (Beta): From Products to Modules]]></title>
            <link>https://medium.com/freshworks-developer-blog/global-apps-framework-beta-from-products-to-modules-74dcf959910f?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/74dcf959910f</guid>
            <category><![CDATA[global-apps]]></category>
            <category><![CDATA[freshworks-marketplace]]></category>
            <category><![CDATA[freshworks]]></category>
            <category><![CDATA[freshworks-developer]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Mon, 11 Mar 2024 11:47:52 GMT</pubDate>
            <atom:updated>2024-03-11T11:47:52.722Z</atom:updated>
            <content:encoded><![CDATA[<h4>A Paradigm Shift in Freshworks App Development</h4><p>Freshworks implemented a multi-product strategy in 2015, allowing developers to create apps for multiple products for their customers, solving business use cases in CX, IT, and CRM industries. The apps would see themselves being part of <a href="https://www.freshworks.com/products/">Freshworks products</a>’ such as Freshdesk, Freshservice, and Freshsales.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k9gcuTPJp-rVgXdoCiINDQ.png" /><figcaption>A sample representation: Modules &lt;&gt; Offerings</figcaption></figure><p>Sooner, we discovered that apps are scoped to specific products, while use cases can span across products. Our solution to this problem is introducing <em>Global Apps.</em></p><h3>Learnings that led to the modular way</h3><p>With platform version 3.0 this year<em>, </em>we launched a new paradigm to build apps on Freshworks — we are calling them <em>modules</em>.</p><blockquote>For developers, a module is a shared unit in their apps that will allow publishing apps to one or more Freshworks product offerings with shared runtime.</blockquote><p>For example, an app they build lists chat_conversation &amp; deal as its modules, and our platform will render the apps for Freshsales and Freshsales Suite users, allowing them to expand their customer reach.</p><blockquote>For customers, a global app can be played across their Freshworks products’ with modules the app has subscribed to.</blockquote><p>As more developers publish global apps, customers will see the apps applying shared context across their products. This will enable cross-product usecases, such as an engineering ticket&#39;s priority, to be automatically set by linking it to an ongoing deal item (of Freshsales) and a support ticket (of Freshdesk) via an app.</p><h3>Modular code interfaces</h3><p>The development experience in building apps on the Freshworks platform becomes <em>modular, </em>starting with <em>platform version 3.0. </em>Consider a case when the app links tickets of Freshdesk and Freshservice to engineering tickets. This common theme of problems demands us to provide newer interfaces for the developers to equip their apps.</p><ul><li>Provide context at the <strong>runtime</strong> about the Freshworks’ product hosting the app. In this case, <em>Freshdesk</em> is where the app is installed.</li><li>Empower developers with local tools for optimal UI and functionality by acknowledging the app’s role in module-based product installations <strong>during development</strong>.</li></ul><p>Developers use the<em> data method</em> to gather contextual data on the current page viewed by the user. For example, a ticket ID of the ticket user viewing — Freshservice ticket? or a Freshdesk ticket?</p><p>The answer is both, depending on the runtime.</p><p><em>Without a modular approach</em>,</p><pre>const data = await client.data.get(&quot;loggedInUser&quot;);<br>// Get the Ticket ID from &#39;data&#39;<br>linkTickets(id); // link with engineering tickets</pre><p><em>With a modular approach</em>,</p><pre>// Know runtime — Freshservice, Freshdesk, or Freshsales?<br>const currentHost = await client.data.get(&quot;currentHost&quot;);<br><br>if (currentHost.subscribed_modules.includes(&quot;support_ticket&quot;)) {<br>    const requesterData = await client.data.get(&quot;requester&quot;);<br>    // link support ticket to engineering sprint<br>}<br><br>if (currentHost.subscribed_modules.includes(&quot;service_ticket&quot;)) {<br>    const = await client.data.get(&quot;user&quot;);<br>    // link service ticket to engineering sprint<br>}</pre><p>Some new features include:</p><ul><li>A <em>currentHost </em>data method that lets the app context about its host for frontend features.</li><li>The custom configuration page expands on <a href="https://github.com/freshworks-developers/global-app-features/tree/main/iparams-html#how-do-i-store-and-retrive-parameters-to-the-platform">_<em>meta </em>to store installation parameters</a>.</li><li>The <em>iparams.json</em> gets <a href="https://github.com/freshworks-developers/global-app-features/tree/main/iparams-json#readme">the modules property</a> to map the store/retrieve operations.</li><li>The changes to <a href="https://github.com/freshworks-developers/global-app-features/blob/main/placeholders/manifest.json">manifest</a> and <a href="https://github.com/freshworks-developers/global-app-features/blob/main/serverless-events/server/server.js">serverless components</a> declare the context of the module and leverage functional reuse.</li></ul><p>Our documentation is in the works and will soon be live on <a href="https://developers.freshworks.com/docs/?tab=modules">the website</a>, along with <a href="https://github.com/topics/global-apps">code samples</a> and <a href="https://developers.freshworks.com/docs/tutorials/">tutorials</a>.</p><h3>Modular local development</h3><p>Every app developer uses Freshworks’ CLI (FDK) to test and ensure bug-free shipping locally. This brings us to the challenge,</p><blockquote>How can we enable developers to test their app on every product module subscribed by the app, instead of just one?</blockquote><p>Developers typically create apps for a specific target audience but then expand to include additional modules and multiple Freshworks product users. Our focus shifted to creating a new setup page that enables developers to switch module runtime on demand while keeping customer views in mind.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J8oH_J7tULF6iAvmH8KRvQ.gif" /><figcaption>Select a module from the manifest to simulate it on Freshdesk or Freshservice (in this case).</figcaption></figure><p>This feature allows developers to easily select the modules they want to test and simulate the tests on the target product. If developers subscribe to modules that support two products, they can test them separately on each product.</p><h3>The Marketplace</h3><p>When developers declare modules in their app, the app will be published to the marketplace and made available to all customers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eH9YlCmZpxo3Ycl0Va8OZw.png" /><figcaption>A screenshot of the selection available for customers to filter through apps</figcaption></figure><ul><li>All the supported apps will be listed in the product app gallery.</li><li>All the apps can be filtered by module or product through the public app gallery.</li></ul><p>The Global Apps Framework has been in a closed beta for a few months. It will soon be generally available in the upcoming months. If you want to be the first to know about this launch, sign up on the community and keep an eye out for our <a href="https://community.freshworks.dev/c/happening/announcements/12">announcements</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74dcf959910f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/freshworks-developer-blog/global-apps-framework-beta-from-products-to-modules-74dcf959910f">Global Apps Framework (Beta): From Products to Modules</a> was originally published in <a href="https://medium.com/freshworks-developer-blog">Freshworks Developer Platform Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Optimizing DevRel for Generalists: A New Team Practice [Part 1]]]></title>
            <link>https://saifshines.medium.com/optimizing-devrel-for-generalists-a-new-team-practice-part-1-b280bd168b16?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/b280bd168b16</guid>
            <category><![CDATA[organization]]></category>
            <category><![CDATA[devrel]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Tue, 28 Nov 2023 02:07:46 GMT</pubDate>
            <atom:updated>2023-11-28T02:07:46.624Z</atom:updated>
            <content:encoded><![CDATA[<p>The evolving developer ecosystem convinced <a href="https://www.freshworks.com/saas/satwik-hebbar-is-proof-passion-binds-life-work-like-nothing-else-blog/">Satwik</a> &amp; <a href="https://www.linkedin.com/in/rohiteddy/?originalSubdomain=in">Rohit</a> that Freshworks needed a developer relations team. They set up one four years ago. Although they had no formal DevRel experience, they would figure things out independently. They took a leap of faith.</p><p>One of the things that kept me excited about the fronting developer experience at Freshworks is its breadth of work opportunities.</p><p>There’s always enough work to:</p><ol><li>Advocacy: Gather developer feedback by influencing product API and platform engineering teams’ design and implementation practices.</li><li>Enablement: Create educational content that helps developers improve their web development skills and advance their careers.</li><li>Experience: Define procedures and standardized engineering documents, PRDs, and code reviews to ensure a consistent and high-quality development experience.</li><li>Community: Foster community within your developer ecosystem by encouraging collaboration and problem-solving.</li><li>Engagement: Motivate developers to contribute, give feedback, and improve their skills by creating incentives and recognition programs.</li><li>Awareness: There are always new platform evolutions for the developers that keep adding value to the developer community</li></ol><p>I present the following picture from <a href="https://twitter.com/mary_grace?s=20">Mary</a>’s blog to articulate the opportunity when hiring new advocates for our team.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*PKjALs7IvC-ALg29.png" /><figcaption>credits to mary_grace</figcaption></figure><h3>Opportunity</h3><p>From a developer’s shoes, the developer experience constitutes the combination of all these segments. It does not matter how we get this experience<strong> </strong>to the developers, but does the ideal experience exist?</p><p>Satwik strongly believed in building a team that complemented each other’s skills so that, as a team, we could deliver the ideal experience to the developers, if not individually. This is amazing because, unlike others in the industry where DevRel is heavily leaning towards either evangelism or advocacy, this approach is the best foundation to leverage balance in the DevRel position.</p><h3>Challenge</h3><p>Grabbing the opportunity means we had to build ways to consume the work routinely to help us build fully-rounded skills and experience. We tried a couple of ways —</p><p><strong>Kanban Board</strong></p><p>The Kanban board on Trello has effectively tracked specific work types but has yet to be more helpful for others. You’ve found that certain swimlanes, including “In Review,” may not be pertinent or useful for particular tasks like program management or ad-hoc assignments. However, the team has discovered that this board format has successfully tracked work like writing blog content or refactoring code.</p><p><strong>Personal Trackers and Standups</strong></p><p>We even took the liberty of counting on the team members to have their projects and responsibilities tracked, done, and shipped in their own ways. The smallest visibility into each other was weekly standups, daily Slack updates, and monthly goal reflections.</p><ul><li>This was helpful for the members who knew how they would achieve their goals. Still, for the team who needed help, there needed to be recurring opportunities to get help by default.</li><li>Our leaders would need help to learn the progress on the projects given to our team, embarrassingly, could have been more significant at posting clear updates daily.</li></ul><p><strong>Freshworks’ Classic Agile (Voluntary)</strong></p><p>I tried cleaning up our internal ticketing system (think Jira). I started setting up tool interfaces best suited to our DevRel and Tech Writing team. However, I did not use authority, nor did I have one to enforce its usage thinking it could not have been a promising approach. A few weeks later, only I and one other were using the system.</p><ul><li>The hope is that team members will adopt it voluntarily, even after gathering feedback or customizing sprints to be monthly.</li><li>Our leaders have the same problem of having difficulty understanding progress unless they specifically ask the team members.</li><li>This approach expected us to have a product manager we did not have during this time.</li></ul><h3>On the Table</h3><p>The more time passes, the more we squander our time.</p><p>The top two contenders were following.</p><ol><li>Comet (Recommended Agile by <a href="https://www.productplan.com/glossary/shape-up-method/">Shape Up</a>) — After a year of trying systems out, it was my best bet for a proposal I want to bring to our DevRel team.</li><li>Enforced Classic Agile — Leverage the existing experience of people in their past organizations and use authority to enforce work in Agile for engineering teams.</li></ol><h3>Enforced Classic Agile</h3><p>To me, it appeared like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/823/0*m9JjrfxGl-cEMZdM.jpeg" /></figure><p>I foresee a few red flags with it:</p><ol><li>To the engineering leaders from the outside, the performance of the DevRel team will be measured by other engineering teams ignoring the nuances. How fair is it?</li><li>It discourages high agency in individuals and paints the picture in black and white; there’s work to do that’s straightforward, or there’s a dependency one wouldn’t care about.</li><li>For example, if DevRel needs a review from the engineering team/would want to review marketing work/would like the product team to prioritize a task, all of them are treated as dependencies or ad-hoc items. The result is to wait until this is someone else’s problem to help you unblock, whereas unblocking yourself through effective collaboration isn’t rewarded.</li><li>Overall, conventional wisdom from the people in authority would start expecting DevRels as “task doers” on fully defined projects (by a product manager). In contrast, unknowns pop up 60% more frequently than usual engineering teams where Design and Scope can permanently be frozen strictly with systems. While this is different for people programs such as champions programs.</li><li>Finally, DevRels proactive focus on collaborating with marketing, engineering, support, and product teams is not rewarded against meeting sprint goals limiting the ability to become partners. Hence constraining the generalist experience Freshworks can offer.</li></ol><h3>Comet (fork of Shape Up)</h3><p>It will take a separate blog for me to expand the details in actionable ways to your DevRel team. I will only share a glimpse of it in this section.</p><p>But here’s how it would appear to you:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/762/0*J2UfHUaEwXiNIvK7.jpeg" /></figure><ol><li>It creates separate projects with necessary talent groups (marketing, engineering) representatives’ commitment upfront, allowing predictability.</li><li>There would be Epic level view for anyone gauging DX, especially to translate work to upper management.</li><li>The Housekeeping and Adhoc work will get their reservations, and PM/EM will have the opportunity to align work to an advocate based on missing areas of the generalist skillset a dev advocate wants to grow into.</li><li>A huge possibility of asynchronous collaboration and decision-making. For example, each task can seamlessly keep every project-level stakeholder, irrespective of talent group, to get inputs and move on without labeling everything a dependency.</li><li>Allows more developer advocates to participate in the product creation process alongside product managers rather than acting as “task doers” on fully defined projects since we assume it can never be fully defined without mutually.</li></ol><p>It is deeply satisfying to think in this direction and re-think existing Practices from 0 -&gt; 1.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b280bd168b16" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Commentary:  2023 State of DevRel Report — Key Insights]]></title>
            <link>https://saifshines.medium.com/commentary-2023-state-of-devrel-report-key-insights-4bb7ab93563a?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/4bb7ab93563a</guid>
            <category><![CDATA[developer-relations]]></category>
            <category><![CDATA[devrel]]></category>
            <category><![CDATA[developer-experience]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Sat, 04 Nov 2023 12:14:33 GMT</pubDate>
            <atom:updated>2023-11-05T04:42:25.949Z</atom:updated>
            <cc:license>https://creativecommons.org/publicdomain/mark/1.0/</cc:license>
            <content:encoded><![CDATA[<h3>Commentary: 2023 State of DevRel Report — Key Insights</h3><h4>Just a bunch of first impressions</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*E8l8H63EVQ85nUCq" /><figcaption>Photo by <a href="https://unsplash.com/@miquel_parera_mila?utm_source=medium&amp;utm_medium=referral">Miquel Parera</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>The <a href="https://www.stateofdeveloperrelations.com/2023devrelreport">State of Developer Relations 2023</a> report is now public. I am happy to see someone is investing their time to curate this knowledge and insights for the internet. Thank you, <a href="https://www.devrel.agency/">DevRel Agency</a>.</p><p>This is and will be an opinionated worldview. I may learn something from you or contribute to your thinking. While reading this, open the report in another browser tab.</p><p>Cheers!</p><h3>The People</h3><h4>Career Paths</h4><p>The age group between 25–44 is about <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=age1">68%</a> of this industry, combined with <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=location">100% of DevRels</a> in India having a DevRel experience between 2–5 years. It implies the <em>lack of adept leaders</em> best qualified to <em>devise clear career paths for DevRel</em> professionals as they grow in India.</p><p>Since over <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=exp">50% of those who have 5–15 years</a> of DevRel experience are from the Bay Area and California, If you live in India, working for companies that offer remote work, reporting, and mentorship opportunities with co-workers at these locations can <strong>enhance</strong> your career.</p><p>The fact that <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=carpath">54%</a> feel a career path needs to be defined proves so.</p><h4>Contracts</h4><p>It is uncommon to see a <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=prof">3.2% to 10.2% growth</a> in a single year for the number of Independent/Freelancers. There are a couple of things we can speculate:</p><ul><li>We will see an uprise in professionals wanting to establish credibility in this space to attract opportunities for themselves.</li><li>Companies find outsourcing tasks more efficient.</li></ul><h4>Degree</h4><p>The report reveals that about <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=education">23%</a> of DevRel professionals did not have a Technical Education. Because Developer Relations has &quot;Developer&quot; in it doesn&#39;t mean one needs to be technical enough to get a DevRel opportunity.</p><p>On the other hand, about <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=res">86%</a> are learning to practice DevRel on the job. It&#39;s interesting to note that many roles in the software industry, such as software engineer, product manager, sales, and marketing, usually have a degree, formal education, courses, or boot camps as learning resources. However, a little formality exists around becoming a Developer Relations (DevRel) professional. Both exciting &amp; worrying.</p><h4>Activities</h4><p>I like how content development is <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=act">split</a> into education (55%) and technical (50%) categories. They indeed demand different skill sets. The education category demands traits such as magnetism, while the other demands the individual to be more illustrative.</p><p>It is exciting to see Developer Experience as a new entrant in 2023 — implying that more and more teams have started to recognize some holistic point of view towards this.</p><p>However, the activities performed <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=act1">at the role level aren&#39;t as insightful</a>. It can be applied similarly to other industries as well.</p><h4>Titles</h4><p>The survey asked people about their titles before joining the DevRel industry. Even after grouping similar titles, there are about <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=toproles">15 groups</a> ranging from Founder to Student. What other industry can be so welcoming?</p><h3>Salaries &amp; Compensation</h3><blockquote>Compensation for Developer Relations practioners continues to <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=atc">rise</a>.</blockquote><p>On and before 2020, the total compensation was consistently hovering at $125,000. Since 2021, it rose to become $225,094 in 2023. While this is a global heuristic, it must apply to India, too, yet the US will remain the primary beneficiary. <strong>That&#39;s an impressive 80.08% increase in 3 years.</strong></p><p>Of all the companies that employed DevRel talent headquartered in the US, only <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=live">51.4%</a> are from the US, and the rest are spread worldwide. Companies in the US know how to harness DevRel talent compared to India.</p><p>The average base salary in India stands at ₹9,16,000 / year over power purchasing parity.</p><blockquote>In companies that run the DevRel program in India, there are 340 DevRels employed (globally) for every 60 companies headquartered in India.</blockquote><p>It&#39;s a comparatively good difference <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=live">compared</a> to any other region. That means every other company (in India) should have at least 6–7 members on their team. If you don&#39;t see that around you, it probably means others are (probably) employed by one among 36% of companies in Silicon Valley.</p><h3>The Companies</h3><p>In a year, the smaller companies (&lt;200 employees) running a DevRel program spiked from <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=size">31.7% to 53.1%</a>. This is a damn good trend. This means experimentation in running these programs rises, and learnings are due a few years later.</p><p><a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=ind">SaaS, Developer Tools, and Cloud Infrastructure</a> are the industries that want DevRel programs the most. If you are a developer advocate, then it is most beneficial for you to add domain expertise in one of these areas.</p><p>Developer First companies have doubled since 2020 from 32% to 62%. There isn&#39;t enough DevRel talent supply. This means companies&#39; growth must demand other functions to accommodate a viewpoint for developer experience. For example, some marketers or engineers are made to see developers as their end users.</p><p>Developer Plus companies have <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=focus">dropped (x2)</a> since 2020 from 67% to 38%. This doesn&#39;t mean these companies have failed because they are developer <strong>plus, </strong>meaning other revenue-generating factors keep the company afloat. This effectively means Developer Plus companies could have done better to run DevRel programs effectively.</p><h3>The Programs</h3><p>The developer community size for both small (&lt;200 employees) and medium (&lt;1000) companies is less than 5000 community members. Optimistically, if the conversation rate adds business value to 5% (global best-considered average), you are on a good trend if you have ~250 monthly active developers.</p><p>Small companies invest ($10k-$50k) more capital into DevRel programs than Medium companies (&lt;$10k).</p><h3>Strategies and Challenges</h3><p><a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=purp">85% of DevRel programs</a> aim to drive awareness and adoption of their products. 76% of them report to marketing (31%), product(26%), and engineering(19%) respectively.</p><p>Yet the <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=topc">top challenges</a> are:</p><ul><li>41% attracting new developers and 39% engaging existing ones.</li><li>44% struggle to create technical content continuously</li><li>32% feel there are limitations with the product they advocate for</li></ul><p>It is evident that an individual needs a different set of skills to solve each challenge, yet the team should be armed with individuals to solve all three. This kind of team demands a leader who can manage these skill sets coming together. Yet, <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=exp">70% have less than 10 years </a>of experience in DevRel.</p><p>Twitter is no longer an effective online channel for DevRels, dropping from 65% to 39% in the same year. It probably means many DevRel teams did not drive enough value from Twitter in one year. Surprisingly, a decentralized social network is rising, currently at 3.4%. It’s <a href="https://joinmastodon.org/">Mastodon</a>.</p><p>Orbit &amp; CommonRoom are new tools that suddenly took the 3rd and 4th spots for DevRel teams to manage their community programs. We were early adoptors of Orbit when its launched. We bet on Orbit, because the philosophy behind it gives us trustworthy lens about the people whiles others obsessed about usage oriented metrics.</p><p>Tremendously worrying aspect — <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=dx">44% of teams</a> never audit their developer experience regarding journey maps or friction logs. Only 1% did monthly.</p><p>All <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=doc">81% of teams</a> think their documentation offering is average. This is instead a good sign, always helping DevRel teams think about how to improve their documentation.</p><h3>The Practice</h3><p>It’s astonishing to see that <a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=challenge">60% of the DevRel teams</a> feel the awareness of the impact created within the organization is the #2 challenge for them. Yet, we don’t see any tactics to solve this problem. This a space for a playbook to emerge, a needed one.</p><p><a href="https://www.stateofdeveloperrelations.com/2023devrelreport?element=recognition">62% believe better-shared practices</a> and measurements with other teams like program management or product is the best-suggested way to drive recognition &amp; credibility for developer relations. This can come as a double-edged sword with grey areas. There must be a better way to think about this.</p><h3>Commentary</h3><p>I went through the reports one after another, started to relate to some of them, and thought of more ideas from the data. So, I began documenting these thoughts in this blog post.</p><p>I appreciate you sticking along and hearing these opinions — of course, held weakly.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4bb7ab93563a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Productive Dissenters in Entrepreneurship]]></title>
            <link>https://saifshines.medium.com/productive-dissenters-in-entrepreneurship-f37841793acb?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/f37841793acb</guid>
            <category><![CDATA[company-culture]]></category>
            <category><![CDATA[enterpreneurship]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Wed, 18 Oct 2023 00:29:46 GMT</pubDate>
            <atom:updated>2023-10-18T00:29:46.043Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zWhGp-QOvY3Q9xUa" /><figcaption>Photo by <a href="https://unsplash.com/@jontyson?utm_source=medium&amp;utm_medium=referral">Jon Tyson</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Entrepreneurs make this mistake over and over and over again. When you start a company, you need help to go far. You want people who believe in you. Belief is the currency. It becomes a lens through which you hire and expand. After you reach a particular scale, this belief may become cloudy. The product or strategy may be perfect, but the belief has become cloudy.</p><p>At this point, some voices surface, pointing out what may go wrong. They do it repeatedly, and you wonder — “Why don’t they see the positive side? Why do they always ‘complain’ the negatives?” — eventually, you push them away from the organization (with all valid reasons).</p><p>What’s unrealized? It’s also the same people who roll up their sleeves, work day and night, and show up daily.</p><p>Here’s the consequence — You surround yourself with people who create a belief echo chamber that makes you blind and furthers you from the ground truth.</p><p>People who repeat all the hints of what could go wrong are productive dissenters. They try to help you until they become helpless. Many entrepreneurs confuse them with negative dissenters who are malevolent.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f37841793acb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Developer Experience 2023]]></title>
            <link>https://saifshines.medium.com/developer-experience-2023-b82a5f868ea8?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/b82a5f868ea8</guid>
            <category><![CDATA[developer-experience]]></category>
            <category><![CDATA[developer-productivity]]></category>
            <category><![CDATA[devrel]]></category>
            <category><![CDATA[developer-relations]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Sat, 14 Oct 2023 06:15:55 GMT</pubDate>
            <atom:updated>2023-10-14T06:15:55.911Z</atom:updated>
            <content:encoded><![CDATA[<h3>Confusion, Opportunity, and a Search for Consensus</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*lLw5g03d4Ud_A9CA" /><figcaption>Photo by <a href="https://unsplash.com/@pagsa_?utm_source=medium&amp;utm_medium=referral">Pablo García Saldaña</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>The developer experience as an industry is undergoing confusion in many ways and is on the verge of acquiring a consensual definition.</p><p>💡The generative AI spawned thousands of startups, all trying to offer developers something to adopt.</p><p>💡 The mid-market and above industry had to transform their offerings powered by generative AI, making the “fast” (development or shipping) ability become a requirement in the software industry.</p><p>💡 As business opportunities emerge in adding value to the developer community, the developer relations industry is starting to see demand before it’s fully matured.</p><p>💡 Every business that started obsessing over KPIs built out like customer support has seen layoffs affect their teams.</p><p>💡 For little less than a decade, every(mostly) DevRel professional has written a blog expressing their day-to-day role repeatedly in their own ways.</p><p>💡More and more companies are implementing developer relations programs in an optimism — reporting into either marketing, ecosystem, product, or engineering.</p><p>💡 Teams from a productivity worldview publish whitepapers to measure experience regarding Flow State and others. Previously, it was DORA, SPACE, and indicators such as Lead time.</p><p>💡The recruitment industry is confused with labeling the right title — Technical Support Engineer, Developer Advocate, Developer Evangelist, Developer Marketer, Developer Relations Engineer, DX Engineer, and much more.</p><p>💡As companies take platform or ecosystem-oriented approaches, new team units labeled Developer Experience Team arose with a combination of DevRel professionals and DX Engineers.</p><p>💡It is now customary for companies to find their first DevRel.</p><p>💡 The emerging products are beginning to see a complex consumer persona — It’s not B2D alone; B2D comes with B2B for the seller to see profitability sooner.</p><p>💡 More and more talent groups outside the Developer Experience industry, like traditional sales, marketing, partnerships, design, and product, are challenged against their skills to serve this technical audience.</p><p>💡The definition of developers spans broader — from engineers within the company to the ones in partnerships at for-profit companies.</p><p>💡 The responsibilities span from Opensource -&gt; B2D -&gt; B2B/B2C, and the nature of DevRel spans from Generalist -&gt; Specialist.</p><p>💡 Technical Writing as a talent equation isn’t nurtured as much as the pace of this industry demands.</p><p>💡 Companies in infrastructure and productivity are looking for DevRel professionals who support internal engineers.</p><p>💡 New startups are now bringing variations of customer service unit equivalents into his industry, such as Developer Success.</p><p>I’m excited to witness this trend and will look forward to it over the years to see how this settles its course — a massive shoutout to everyone in this space for making this a forward-moving industry.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b82a5f868ea8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Celebrate this Hacktoberfest with Freshworks]]></title>
            <link>https://medium.com/freshworks-developer-blog/celebrate-this-hacktoberfest-with-freshworks-1c8497bfb961?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/1c8497bfb961</guid>
            <category><![CDATA[announcements]]></category>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[hacktoberfest]]></category>
            <category><![CDATA[freshworks-developer]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Fri, 09 Oct 2020 11:12:51 GMT</pubDate>
            <atom:updated>2020-10-09T11:12:51.260Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lij-jwozNniAEGnHGvIQ8w.png" /></figure><p>A celebration of open-source returns this year with<a href="https://hacktoberfest.digitalocean.com/"> the announcement of Hacktoberfest</a>. At Freshworks, we took the opportunity for the developer community to explore building Freshworks Apps that can power customer &amp; employee centric teams.</p><p>As beginners, we have always learnt programming by building a To-Do list App, a Pomodoro App or something similarly small but purposeful.</p><p>While that’s a good start, not every beginner manages to be inspired beyond that. So they search for inspiration on the internet only to end up with more ideas that are very similar. How about exploring,</p><ol><li>An app that will remind a support representative to take a break every 60 minutes?</li><li>An app to show a sales representative the details of the product exactly when the customer needs to learn about it?</li></ol><p>The same Pomodoro app you were inspired to build can be tweaked to help a support person take a refreshing break every 25 min and the same to-do app can help a sales representative remember all the details that a customer needs.</p><blockquote>There is always a pool of problems waiting to be solved by you.</blockquote><h3>A Freshworks App</h3><p>Technically, a Freshworks app is a regular web-app but completely hosted, run and scaled by the developer platform. Every app that runs on the Freshworks platform is built <a href="https://developers.freshworks.com/documentation">with SDK and API tools</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/880/0*FnUZBa0JO5al1SYu.jpg" /><figcaption>An Freshworks App (on Freshdesk) showing attachments within a long thread in a ticket.</figcaption></figure><p>You are looking at a Freshworks app in the bottom right corner of the image above. This app runs within the Freshdesk web application and displays all the attachments posted in a conversation in a specific location without having to scroll through a long conversation. It saves time, doesn’t it?</p><p>We have compiled <a href="https://developers.freshworks.com/tutorials/">tutorials</a> which you can learn from to build an app. They are step by step to let you learn at your own pace.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2I2XLBWLVocRTm-8kBJ1jg.png" /><figcaption>A snapshot of available tutorials from Freshworks code recipes.</figcaption></figure><h3>Hacktoberfest</h3><p><a href="https://hacktoberfest.digitalocean.com/">Hacktoberfest 2020</a> is a month long virtual event to celebrate open source contributions presented by <a href="https://www.digitalocean.com/">Digital Ocean</a>, <a href="https://hacktoberfest.digitalocean.com/intel.pdf">Intel</a> and <a href="https://www.dev.to/">DEV</a>.</p><p><a href="https://hacktoberfest.digitalocean.com/">Hacktoberfest &#39;21</a></p><p>During the entire month of October 2020, all you have to do is to contribute to any open source project and create at least <strong>four</strong> valid pull requests. It could be a bug fix, enhancement, or even a documentation change!</p><p>Also, to familiarize yourselves with business problems as they exist today, we created a <a href="https://github.com/freshworks/marketplace-sample-apps/issues">few tasks</a> to steer you as to where you can invest your time.</p><p>You can explore two of our Open-source repositories to start with,</p><h4>Crayons</h4><p><a href="https://github.com/freshworks/crayons/issues">Crayons</a> is Freshworks’ UI component Library. It offers components for developers to build any Frontend view that adapts to the Freshworks Design system. We are positively vulnerable. We might be missing a component, or there could be a bug in a component. Maybe one of them is not accessible enough or might be having unclear documentation. Raise an Issue or a PR and become a part of our growing community!</p><p>Check the <a href="https://github.com/freshworks/crayons">Github open source repository</a> of sample apps for contributions.</p><h4>Sample apps</h4><p>We’ve all been there. When docs fail to help you, sample code is what we all start looking for. So, this Hacktoberfest you could:</p><ul><li>Try <a href="https://dev.to/saif_shines/try-something-different-this-hacktoberfest-may-be-a-freshworks-app-53cj">getting started</a> or jump straight <a href="https://github.com/freshworks/marketplace-sample-apps/issues">into our Issues</a> to see if you can find something and get familiar. I promise most of them would be fairly easy.</li><li>If you are a bit experienced with Freshworks App development, I am sure you will find tons of issues with the code. We encourage you to report them or raise a PR.</li></ul><p>Check the <a href="https://github.com/freshworks/marketplace-sample-apps">Github open source repository</a> of sample apps for contributions.</p><blockquote><em>Try something different this Hacktoberfest. Maybe a Freshworks App?</em></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1c8497bfb961" width="1" height="1" alt=""><hr><p><a href="https://medium.com/freshworks-developer-blog/celebrate-this-hacktoberfest-with-freshworks-1c8497bfb961">Celebrate this Hacktoberfest with Freshworks</a> was originally published in <a href="https://medium.com/freshworks-developer-blog">Freshworks Developer Platform Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Ideal Hackathon motivation]]></title>
            <link>https://saifshines.medium.com/the-ideal-hackathon-motivation-1304b0f5c589?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/1304b0f5c589</guid>
            <category><![CDATA[hackathon-organizing]]></category>
            <category><![CDATA[hackathons]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Sat, 15 Feb 2020 07:09:36 GMT</pubDate>
            <atom:updated>2020-02-15T13:37:34.278Z</atom:updated>
            <content:encoded><![CDATA[<h4>An Ideal motivation guide for hackathon managers, participants and judges.</h4><p>There are lots of guides/articles (like <a href="https://www.hackerearth.com/community-hackathons/resources/e-books/guide-to-organize-hackathon/">this</a> &amp; <a href="https://hackathon.guide/">this</a>) that tell you everything related to hackathon. Usually, the motivation that drives these processes is fuzzy. I want to make that clear by telling you flaws those occur without clear motive. So, most of the hackathon stories turn out this way.</p><ol><li>Organisers reach out companies/coworking spaces for venue. They promise some promotion benefits from the attendees and get a venue.</li><li>Organisers reach out some popular candidates from some of their connections and invite them as a Judges.</li><li>Participants see the reward of attractive cash prizes and participate to win the cash prize. If they don’t they get away saying — “I learnt something”</li><li>They call all the participants and work on for 24 to 48 hrs to build a project.</li><li>A Group of 5 judges will rate the projects on some parameters.</li><li>The average of the scores decides the best project that wins.</li></ol><p>That is the story painted. Then, everyone disperse and never see each other again. There is nothing wrong with this painting that is being practiced because either ways all the parties are satisfied because something good is happening.</p><blockquote>“To power the trajectory of a project is what hackathon is meant to fuel for. This is getting diluted. With enough people practicing the same story repeated;wrong precedence is set for people who look forward to host more hackathons.”</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*wsP0xmQjAY_SdLD1" /><figcaption>Photo by <a href="https://unsplash.com/@spacex?utm_source=medium&amp;utm_medium=referral">SpaceX</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Journey</h3><p><em>Identifying a problem </em>is the <strong>starting point of trajectory</strong>. <em>Trying to build a solution</em> is <strong>the path innovator decides</strong>. <em>Making this solution available for everyone in the world </em>to use is the <strong>goal</strong>(<em>which requires it turn into org</em>).</p><blockquote>Hackathon is the thruster powering that journey ↖</blockquote><h3>What’s flawed?</h3><p>This thruster.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*o8Mc8YSyEiGhzSgr" /><figcaption>Photo by <a href="https://unsplash.com/@meric?utm_source=medium&amp;utm_medium=referral">Meriç Dağlı</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>It is built on top of combined motive of Organiser, Participant &amp; Judges.</p><ul><li>Motivation as a organiser to host the hackathon</li><li>Motivation as a participant to build a solution</li><li>Motivation as a Judge to give constructive feedback.</li></ul><h3>Hackathon manger’s Flaw</h3><p>Hackathon managers are busy bugs. They have a lot things to take care of. It’s not because they are managers. It’s because they failed to build the right organising team and tools. So they don’t have bandwidth for focus on invisible yet most important aspects.</p><p>Hackathon is very <a href="https://en.wikipedia.org/wiki/Hackathon">broad</a> term. This ideally allows all the possible industries(<em>finance, cloud, blockchain, so on</em>) to be invited.</p><ul><li>Each industry has different degrees of complexity in Business model, Technology substance, Marketing strategy.</li><li><em>Usually</em>, each of Judges can only be expert in <strong>one industry</strong> + any of <strong>two degrees of complexity</strong>. So they are not reliable on everything just because of their work title.</li><li>This is why hackathons need <a href="https://www.hackathon.com/theme">themes</a>. Having judges to specific themes paves the way for fair process of judging.</li><li>But still blindly adding up scores across the projects is unfair. Have the debate process at the end where one Judge from each theme should back the best project. A constructive argument will give a fair winner.</li><li>The Judging criteria comes from scope of Hackathon. For example, having “Is this project capable to having potential IPO” in a college level hackathon is very stupid without product going to the market with private investment. Take care when you build this Judging criteria.</li><li>Number of teams participated is only indicator of good promotions not for the hackathon. Hackathon is just a thruster; it only gives teams a push and best amongst a voice of recognition.</li></ul><h3>Participant’s Flaw</h3><p>Let me put this straight. You are a rocket heading to serve thousands with the solution that you’ve come up to solve a problem.</p><ul><li>Before cash prize or any reward, you primary goal should be improvement of project you built. So hear others thoughts even if they are stupid. They’ll say what people think about you. Find and fix what’s causing them to think that.</li><li>Ask Judge the questions. Phrase your question to get as much feedback as possible. There’s only a thin line until it converts to argument. Once you argue, you lose the feedback. Doesn’t matter you win $10,000; if you don’t fix from what you hear.</li><li>If you are winner of Hackathon, the organising team does over marketing of your success. In reality, you just won amongst couple of teams (20–50) participated. Real world is different.</li></ul><h4>Pitch</h4><ol><li>This is my team and background. This is how long We’ve been working on this.</li><li>This is the problem we are trying to solve.</li><li>This is how we build app/website/device using this particular technology.</li><li>Here is the data about these many people having this problem. Potentially, they may buy. Prove it with math.</li></ol><p>Let Judges ask you questions about what is unique, why particular price, why particular market, what’s different in the algorithm, competitors, so on. You just be prepared with enough research to answer them.</p><p>Don’t make answer on spot. Be free to say it’s own assumption and you can work on it better.</p><h3>Judging Flaw</h3><p>You have an opportunity to do important activity of this thruster. To help rocket stay in trajectory.</p><ul><li>Incase you are given Judging criteria. Strictly abide by it. It is common bar for all the participants. Don’t be biased on basis of presentation or communication skills. Content they deliver is priority.</li><li>Be candid. Ask them if they have any questions for you. Answer them if you could be of any help to them.</li><li>Tell them you don’t know things you don’t know. Inturn ask them, you get to learn.</li><li>Don’t try to rate every thing in the Judging criteria sheet if you understand you are not from <em>relevant industry</em> or <em>expert in particular degree of complexity</em>. Rate what you think + add comments. Feel free to discuss for help with other judges alongside you at stage of finalizing.</li></ul><blockquote>These are flaws to be fixed. It is important. Share it to Judges, Participants or Hackathon managers. I want you to set the right precedence for future.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1304b0f5c589" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Get Certified! — Freshworks Developer Certification Program]]></title>
            <link>https://medium.com/freshworks-developer-blog/get-certified-freshworks-developer-certification-program-ebda6a65d485?source=rss-6eeb07f34fec------2</link>
            <guid isPermaLink="false">https://medium.com/p/ebda6a65d485</guid>
            <category><![CDATA[app-development]]></category>
            <category><![CDATA[developer-certification]]></category>
            <category><![CDATA[announcements]]></category>
            <dc:creator><![CDATA[Saif Ali Shaik]]></dc:creator>
            <pubDate>Thu, 12 Dec 2019 06:03:52 GMT</pubDate>
            <atom:updated>2019-12-12T06:03:52.527Z</atom:updated>
            <content:encoded><![CDATA[<h3>Announcing — Freshworks Developer Certification Program</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-eo-tMx5dpZ5PvL-cvy9RA.png" /></figure><p>Over 150,000 businesses rely on Freshworks to help them win customers for life. These businesses tailor the experiences delivered by our products to match their individual preferences.</p><p>Today, we have close to a 1000 marketplace apps and over 2000 custom apps built using the Freshworks SDK and APIs, on our developer platform.</p><p>This means,</p><ul><li>Businesses are now actively seeking the help of developers to solve complex business problems.</li><li>Developers are leveraging the huge Freshworks customer base to deliver the next app success story.</li></ul><h3>Milestone is set</h3><p>Generating these two opportunities was a decisive milestone for us. This journey encompassed analysis of numerous business use-cases, building a platform rich with features to address these use-cases, as well as significant effort spent on app reviews to deliver a marketplace with high-quality apps.</p><p>Businesses rely on apps significantly. Their ability to win customers for life is proportional to the quality of marketplace apps they choose to use or custom apps they choose to build. The role of skilled developers in building these apps is invaluable and a certification program can be a great validation for developers themselves or the businesses that work with these developers. Developers who get certified will be part of our trusted partner network and attract project opportunities from our customers.</p><h3>Devising the certification program</h3><p>Keeping the intent in perspective, when working on the certification program, we needed to come up with an efficient, repeatable, and fair process. The scalable process we ended up with involves collecting questions that then go through the following review cycle:</p><ol><li><strong>Technical Review</strong> — Is the question technically correct?</li><li><strong>Ambiguity Review </strong>— Is the question unambiguous? Did authors miss looking at the question from a different paradigm?</li><li><strong>Content Review</strong> — Does the question follow an easy-to-understand terminology?</li></ol><p>Developers can expect questions based on Web Development, App Security, Serverless App Development, Debugging, and the Freshworks SDK in general. The certification is designed to test a developer’s knowledge along the breadth and depth of the developer platform:</p><p><strong>Breadth</strong> — Awareness of the basic and advanced features of the platform and how apps work.</p><p><strong>Depth</strong> — Awareness about the specific workings and use-cases of each feature.</p><p>We believe that our developer certification program will help us in our endeavour to connect our best developers with our most demanding business opportunities. We already have our first few developers certified during our last two <a href="https://medium.com/freshworks-developer-blog/recap-freshhuddle-berlin-722c80fe6870">Freshhuddle developer events</a>.</p><p>Watch this space for certification opportunities near where you work.</p><p>You can always write to us at <a href="/service/mailto:marketplace@freshworks.com">marketplace@freshworks.com</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ebda6a65d485" width="1" height="1" alt=""><hr><p><a href="https://medium.com/freshworks-developer-blog/get-certified-freshworks-developer-certification-program-ebda6a65d485">Get Certified! — Freshworks Developer Certification Program</a> was originally published in <a href="https://medium.com/freshworks-developer-blog">Freshworks Developer Platform Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>