<?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 Shmuel Disraeli on Medium]]></title>
        <description><![CDATA[Stories by Shmuel Disraeli on Medium]]></description>
        <link>https://medium.com/@shmuel-disraeli?source=rss-175e1c1f3467------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*M48t5Hc_8bYWd7LSHlQIkQ.png</url>
            <title>Stories by Shmuel Disraeli on Medium</title>
            <link>https://medium.com/@shmuel-disraeli?source=rss-175e1c1f3467------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 04:10:37 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@shmuel-disraeli/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[The Individual Contributor Role Is Changing]]></title>
            <link>https://medium.com/israeli-tech-radar/the-individual-contributor-role-is-changing-effd8df4b388?source=rss-175e1c1f3467------2</link>
            <guid isPermaLink="false">https://medium.com/p/effd8df4b388</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[software-architecture]]></category>
            <category><![CDATA[tech-lead]]></category>
            <category><![CDATA[clean-code]]></category>
            <category><![CDATA[coding-agents]]></category>
            <dc:creator><![CDATA[Shmuel Disraeli]]></dc:creator>
            <pubDate>Sun, 04 Jan 2026 18:36:12 GMT</pubDate>
            <atom:updated>2026-01-04T22:28:59.776Z</atom:updated>
            <content:encoded><![CDATA[<p>For a long time, the Individual Contributor (IC) role was easy to define:<br>You wrote code, reviewed code, and shipped features.</p><p>That definition no longer holds.</p><p>Over the past year, working with teams adopting AI coding agents, I’ve seen the IC role shift in a fundamental way. The change became very clear to me during a recent project.</p><h3>What Happened on the Project</h3><p>At one of my recent clients a cyber-security company, I helped design a new React frontend as part of a transition from an on-prem product to a SaaS architecture. <br>Alongside the architectural work, I trained developers who were strong backend engineers, and were joining the frontend effort.</p><p>What surprised me most was not the speed of their delivery, but <strong>where the real impact came from</strong>.</p><p>The most impactful thing I did on that project was not implementing the client side architecture or even training the devs on the new project. It was defining <strong>rules, guidelines, and principles for AI code agents to follow</strong> inside the codebase.</p><p>Once those rules were in place, two developers with no prior React experience became capable, confident contributors.<br> <br>The AI didn’t replace learning but it dramatically accelerated it when it operated within clear constraints.</p><h3>Some Key Insights</h3><h4>The IC Is Now the AI’s Team Lead, QA, and DevOps</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*x212KOq0iABsD6JzHxXfcg.png" /><figcaption>you are coding the rules and tools for the machine that codes</figcaption></figure><p>In practice, our role shifted.</p><p>we still write code, but more importantly we define tasks and context for the AI, <br>provide <a href="https://modelcontextprotocol.io/">MCP’s</a> and privileges for tools, <br>reviewe AI-generated code<br>validate the AI’s work,<br>last but not least we write rules and principles</p><p>The IC is no longer just an executor. The IC is now the <strong>AI’s team lead, QA, and DevOps engineer </strong>responsible for quality, safety, and direction.</p><h4>Time Estimates Have Fundamentally Changed for the better with a twist</h4><p>This shift had a dramatic effect on time estimates.</p><p>A task like implementing a modal window in the design system used to take a capable frontend developer 2–4 hours. With AI and clear guidelines, it became a ~20-minute task even for developers new to frontend.</p><p>but here is the twist <br>the actual time is spent on guiding the AI through the task and checking the output, the actual coding by the AI takes just a few minutes</p><h4>Code quality tools Are Now the Standard and There’s Room for More of them</h4><p>Tools like TypeScript and various testing frameworks have moved from “best practice” to <strong>baseline infrastructure</strong>.</p><p>In the past, teams debated whether types and tests slowed development. With AI writing code, that tradeoff largely disappears. AI generates types and tests fast, while the feedback they provide becomes critical. and saves time in the validation process.</p><p>Types and tests now serve a dual purpose, <br>they provide fast, unambiguous feedback to AI,<br>and they help the IC validate the AI’s work faster.</p><p>Once you accept that AI is a first-class contributor, it becomes obvious that we need even more tools that produce <strong>fast deterministic signals about correctness and intent</strong>.</p><h3>The IC Is Becoming AI Oversight</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bGOASmmHkSHcDIX3Lb9tUA.png" /></figure><p>Taken together, this leads to a broader conclusion.</p><p>The Individual Contributor role is evolving into <strong>AI oversight</strong>.</p><p>As AI models and tools evolve and become more and more capable The IC’s value increasingly comes from:</p><ul><li>defining constraints</li><li>shaping feedback loops</li><li>owning and signing off work produced by AI agents</li><li>being the missing accountability layer for AI</li></ul><h3>Tech debt does not disappear with AI, It can even accelerate unless actively governed.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GgjVZKxyFdfBgbAdKGz3Zg.png" /></figure><p>as AI is generating code fast, and IC’s are reviewing and screening out more and more garbage produced by AI, even with good rules and guides that refine the AI’s outputs,<br>a lot more code is produced and before you notice you need to maintain a huge project with lots of interconnected features, and the AI is lousy at refactoring, as this requires a broader context of the system and maybe even the company and so you still need someone to look at the system more broadly.</p><p>the tech lead role is becoming <strong>AI governor</strong> setting up the tools, rules, and policies that determine how AI operates,<br>to make sure faster code does not mean low quality unmaintainable code<br>AI would not suggest meaningful code refactors, or adding tests prior to making a code change and so on…</p><p>its on the tech lead to make sure the team can move fast without accumulating lots of code that would be hard to refactor or even maintain</p><h3>One Last Critical Insight: AI Doesn’t Stop to Ask</h3><p>There’s a final insight that deserves its own focus.</p><p>AI does not naturally slow down or ask clarifying questions. It will confidently start coding in ambiguous areas places where even a senior developer would hesitate to touch without deeper business context.</p><p>This makes AI agents extremely risky.</p><p>Introducing AI into a codebase requires mapping the system and explicitly signaling:</p><ul><li>where AI can move fast,</li><li>where it should slow down,</li><li>and which areas are too risky to touch without human involvement.</li></ul><h3>What’s Next</h3><p>In the next post, I’ll go deeper into how this oversight can actually be implemented in practice:</p><ul><li>how to identify risky areas in a codebase,</li><li>how to communicate those risks to AI,</li><li>and how tooling can help control speed without killing momentum.</li></ul><p>Because if AI is now a first-class contributor, we need systems designed for that reality.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=effd8df4b388" width="1" height="1" alt=""><hr><p><a href="https://medium.com/israeli-tech-radar/the-individual-contributor-role-is-changing-effd8df4b388">The Individual Contributor Role Is Changing</a> was originally published in <a href="https://medium.com/israeli-tech-radar">Israeli Tech Radar</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Real Problems of the crypto ecosystem]]></title>
            <link>https://shmuel-disraeli.medium.com/the-real-problems-of-the-crypto-ecosystem-5d50ce12365?source=rss-175e1c1f3467------2</link>
            <guid isPermaLink="false">https://medium.com/p/5d50ce12365</guid>
            <category><![CDATA[decentralization]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Shmuel Disraeli]]></dc:creator>
            <pubDate>Fri, 18 Jun 2021 08:11:52 GMT</pubDate>
            <atom:updated>2021-06-21T02:41:32.207Z</atom:updated>
            <content:encoded><![CDATA[<p>In the previous article called <a href="https://shmuel-disraeli.medium.com/why-do-we-even-need-crypto-29468e809454">why do we even need crypto</a>, we talked about how crypto could solve the worlds problems, <br>but before crypto could do that, the crypto Eco system needs to solve its own problems first.</p><p>In this article I&#39;m going to name a few major issues and try to suggest an abstract solution</p><p>The first one that comes to mind is the high level of complexity in crypto projects, this is leaving a lot of people out, I talked about this problem in a previous article called <a href="https://shmuel-disraeli.medium.com/simplicity-221676c697b0">simplicity</a>,</p><p>It&#39;s hard to reason about a complex problem, in order to address complex problems in a simple way you need better building blocks, developers call them layers of abstraction, Think how hard it would be for you to explain what a house is without having a way to talk about its various parts like roof flor windos door but instead explaining it in the low level raw matirials layer like sand wood metal stone,<br>It would be close to impossible,<br>this is where crypto is today.</p><p>But for good abstractions to form we need a larger ecosystem and with the quantity we would also get a decent amount of quality,</p><p><em>For a few years now the largest developer community is in JavaScript.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vzxRVkd32X3Wzr1gDQQLGA.jpeg" /></figure><p><em>You could also check out these </em><a href="https://madnight.github.io/githut/#/pull_requests/2021/1"><em>Github</em></a><em> and </em><a href="https://insights.stackoverflow.com/trends?tags=javascript%2Cpython%2Cjava%2Csolidity"><em>Stackoverflow</em></a><em> stats links</em></p><p>So for the new global human cooperation that crypto is promising us to succeed we will need a lot more developers to get in the field.</p><p>Smart contracts introduce this concept that “code is law”<br>Well if code is the law then we need a lot of people that can understand the law</p><p>same way as literacy is the fundamental requirement from citizens to be able to understand what the government is doing and to be able to vote</p><blockquote><em>“Information is the currency of democracy”</em> (<a href="https://www.quotes.net/extsearch.php?src=amazon&amp;s=Thomas+Jefferson&amp;dep=Books">Thomas Jefferson</a>)</blockquote><p>I must be fair and say that Ethereum Smart contracts can be written in Solidity which tries to resemble JavaScript or in another flavor called Viper which is trying to be similar to Python, obviously, Ethereum recognizes the issue and is trying to address it,<br>In reality, smart contract development still has a steep learning curve.</p><p>The second issue I want to address is that a big part of crypto is distributed computing architecture and networking, which is not a common ground for a lot of developers today.<br>Also, these networks run on significantly stronger computers than your everyday laptop, it is said that it would take a normal laptop 10,000 years to compute Bitcoins proof of work that happens every 10 minutes to produce a new block.</p><p>To be fair some blockchains try to address it to allow maximum participation from everyday users, with normal computers, using various techniques, but same as before there is still a high barrier to entry and unfair advantages towards Optimised computers participating in the network.</p><p>The reason this is important is that the same way that software development barrier to entry to blockchains is high hardware network participation is also high, and this centralizes the computing power in the hands of the few, which can cause censorship and other issues with people abusing their power… <br>Those very issues that blockchains aim to solve in the first place.</p><p>like I said before crypto can decentralize the world, we need to decentralize crypto.</p><p>So how do we solve these two complex issues<br>1. Finding a popular programming language that would include the largest amount of developers in the crypto ecosystem.</p><p>2. Finding a distributed computing and networking architecture that is simple enough to allow maximum participation.</p><p>There&#39;s a known fact that a big part of our accelerated technological advancement in the last hundred years was caused by the ability to build upon existing infrastructure,<br>For example, the first electrical products were hooking to the lightbulb socket, they didn&#39;t use the outlets we use today, We used to use telephone lines to connect to the Internet,<br>so as you see major innovations in the modern era we&#39;re riding on existing infrastructure.</p><p>Maybe we can find a way that crypto could ride on existing programming languages existing code runtime environments, existing infrastructure, and networking, maybe something that is already installed in every household (hint hint)</p><h3>Eureka! the Browser</h3><p>What if we could run smart contracts in the browser using JavaScript,<br>In such a way that every person with a computer and an Internet connection could run smart contracts &amp; participate in the network,</p><p>and on the software side, almost every web developer in the world could read and write smart contracts, and develop new solutions with his existing toolkit.</p><p>I know what you&#39;re thinking, <br>This is an ingenious idea, but It will never work, <br>In reality what would stop someone with an optimized computer farm or a data center to run a lot of “browsers” and get an unfair advantage over the home user?</p><p>And that is exactly what we are going to talk about in the next article…</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5d50ce12365" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why do we even need Crypto?]]></title>
            <link>https://shmuel-disraeli.medium.com/why-do-we-even-need-crypto-29468e809454?source=rss-175e1c1f3467------2</link>
            <guid isPermaLink="false">https://medium.com/p/29468e809454</guid>
            <category><![CDATA[decentralization]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[crypto]]></category>
            <dc:creator><![CDATA[Shmuel Disraeli]]></dc:creator>
            <pubDate>Wed, 16 Jun 2021 13:09:19 GMT</pubDate>
            <atom:updated>2021-06-16T13:13:38.462Z</atom:updated>
            <content:encoded><![CDATA[<p>So after the previous article about the need for simplicity, I would like to step back a bit and think about crypto from first principles.</p><p>why do we even need crypto?<br>And when I say crypto I mean a decentralized currency<br>I&#39;ll try to do it in a problem solution Q&amp;A format.</p><p>What does crypto allow us on the most basic level?</p><p>Crypto allows us to reach a consensus about the state of a Ledger, this allows us to create a currency that incentivizes the security of that ledger so it&#39;s circular.</p><p>Let&#39;s go in a bit deeper,<br>why do we need a currency?</p><p>The main reason is the ability to do trade without direct barter because barter requires that you would always have a valued good at your disposal to trade for stuff you need.</p><p>let&#39;s go even deeper why do we need barter???</p><p>Hang in there… :) <br>Because it’s difficult for a single individual to get all the basic necessities by himself and that is also why humans live in a community.</p><p>This also explains why communities are forming around cryptocurrency&#39;s</p><p>We reached the final principle humans need to cooperate in a community for food security and shelter, but what if a small community is able to provide for themselves why would they need a currency?</p><p>Much like crypto itself, the answer is a circular game theory, <br>the basic human family can provide food shelter, and security witout a currency until it is faced with a tribe that is stronger, <br>then they have to form into a tribe that can provide food security and shelter, <br>but then they are faced with a bigger tribe, and they compete for water or territory to grow food, <br>so now they need a way to protect themselves from the bigger tribe so they make an alliance with another small tribe, but then they are faced with war from a bigger alliance they need to form a bigger structure something that looks like a nation with a king and an army, then the nation is threatened by another nation, this starts a race to the large unions we see today like the United States, Soviet Union or the European Union that are spreading across whole continents, to protect from an invading empire.</p><p>With every restructure of Humane society, <br>Society needs to find better means for cooperation, <br>And as the game theory race above shows we are headed to a single world nation spread across the Internet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nHPMkn9bdzgE4s3fPjPC7A.jpeg" /></figure><p>After going deep into the first principles of why a currency is needed let&#39;s try to answer the original question why do we need a decentralized currency?</p><p>On the first block of Bitcoin, this message was embedded</p><blockquote>The Times Jan/03/2009 Chancellor on brink of second bailout for banks</blockquote><p>now we can fully understand it, <br>you can read about the 2008 crisis, there are books and movies.<br>But in my opinion, the bailout mostly shows where the governments are failing,<br>as we said the basic reason that we have a currency system is to provide security food, and shelter for the individual, what happened in the bailout is the exact opposite people lost their homes their savings because banks abuse their power and the government instead of helping the injured Poor people they helped the strong powerful banks the ones causing the whole mess in the first place.</p><p>so we see that on large scale governments and currency systems fail to solve the basic problem that is the reason for their existence…</p><p>so for humanity to scale to a global society working together, we would need a better financial system that individuals could trust in, a system that would provide for them the security food, and shelter they need.</p><p>And this can only be achieved by decentralizing the power of banks governments &amp; large corporations, there is a reason why humanity chose different types of democratic government models over kings &amp; dictators because on a big scale they accumulate too much power and they can abuse it in horrifying ways,<br>The same goes for governments and corporations on the scale of continents they accumulate too much power.</p><p>so humanity is moving in two opposite directions <br>1) is becoming one global society that makes us all use the same centralized services ($USD, Amazon, Facebook, Google, Chinese manufacturing, Big tech, Big Pharma)<br>2) decentralizing power from organizations and individuals that accumulates too much power (kings, empires, monopolies, cartels)</p><p>Crypto can save us by incentivizing decentralization.</p><p>But before crypto would be able to solve humanity&#39;s problems, the crypto ecosystem needs to solve its own problems and that is exactly what we&#39;re going to talk about in the next article.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=29468e809454" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Simplicity]]></title>
            <link>https://shmuel-disraeli.medium.com/simplicity-221676c697b0?source=rss-175e1c1f3467------2</link>
            <guid isPermaLink="false">https://medium.com/p/221676c697b0</guid>
            <category><![CDATA[blockchain-development]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[decentralization]]></category>
            <category><![CDATA[crypto]]></category>
            <dc:creator><![CDATA[Shmuel Disraeli]]></dc:creator>
            <pubDate>Mon, 24 May 2021 04:32:59 GMT</pubDate>
            <atom:updated>2021-06-01T19:29:16.074Z</atom:updated>
            <content:encoded><![CDATA[<p>I have prepared a technical article with a mapping of all the current problems in the crypto ecosystem, And what it will take to push the whole ecosystem forward, In it, I refer to technical things that are difficult to understand and explain.</p><p>but I could not finish it nor did I feel that I would want to publish it…</p><p>My heart didn’t let me and as you know the heart feels things that the mind and mouth are not yet able to express…</p><p>I started thinking about why I do not want to publish it… Then It hit me</p><p>It’s just too long and complex</p><p>There is a famous saying</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/282/1*QbDrTRuRf30c3d3G6HQb5Q.png" /></figure><blockquote>“being simple is not simple”</blockquote><p>And this is much more noticeable in the crypto world where even really smart people can feel they are getting lost in complexity &amp; things are getting senseless.</p><p><strong>Why are crypto projects so complex to explain?</strong></p><p>The answer lies in the main problem that Bitcoin has tried to solve and in the broad surface for solving the problem.</p><p>Usually, a project tries to solve a specific problem with the help of technological innovation</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/197/1*DtN7kfe28Nrxf1k5rjdpHw.png" /></figure><blockquote>“our innovation will save you time or money”</blockquote><p>Then things make sense and are relatively simple</p><p>There is a problem within a specific domain and there is a solution within that domain <br>using some innovation.</p><p>But in crypto, it’s not like that</p><p>There is a problem in the world of economy / society / software / internet / distributed computing infrastructure / privacy / government / information security</p><p>And there is some crypto project trying to solve it</p><p>with some innovation in these fields…</p><ul><li>software</li><li>Encryption</li><li>Mathematics</li><li>economy</li><li>game theory</li><li>philosophy</li><li>Computing infrastructure</li><li>A new method of government</li><li>&amp; more…</li></ul><p>Crypto has a wider surface of problems and a wider surface for problem-solving which in turn introduces new problems</p><p>And as a result, one can get lost</p><p>You’re having a conversation with someone in the field and the conversation can move without noticing From Computer Infrastructure to Economics to Philosophy to Game Theory to Computer Science Back to Philosophy….</p><p>There is an internal joke that many developers in the field will identify with</p><blockquote>Crypto will solve the world’s problems, yeah right</blockquote><p>On the one hand, the technology and the ideas it presents have tremendous potential and anyone can feel it.</p><p>On the other hand, the technology introduces so many new complexities that it can barely solve its own problems, <br>recently we started hearing people talk about Centralized decentralized Finance CeDefi</p><p>How can it be both decentralized and centralized? And what is the value of cryptocurrencies if they are managed on a private system?!</p><p>In short, it is a mess,<br>and I know some of you would have arguments for this approach (which is kind of opposite to the initial Bitcoin white paper)</p><p>but in my opinion, this is a big signal from the ecosystem that we are trying to work around the complexity we just have <a href="http://www.paulgraham.com/schlep.html"><strong>schlep blindness </strong></a>to it.</p><p>The hidden and most important feature in any system and in any organization is simplicity</p><p>Why simplicity because simplicity pays off</p><p>Simplicity for users to interact with, simplicity for developers to develop, simplicity for marketers to explain, and sales to sell. <br>Simplicity helps unite everyone around the same idea,</p><p>a few years ago I had the privilege of working with a senior software consultant on an interesting project in AdTech that included working on server infrastructure and also some client-side</p><p>we worked on all sorts of things around the stability of our servers because they were supposed to withstand large volumes of traffic, and we launched fast not really ready for the load this was before DevOps was a real buzzword, and my manager asked me to take advantage of the hours with the consultant to push some new features in “by the way” …<br>when I tried to do that</p><p>the consultant Stood up and said</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/86/1*k4EuzYdJn7J9fYt4kAo40A.jpeg" /></figure><blockquote>“the only new feature that will get in this system is stability!”</blockquote><p>This is exactly how I feel about the world of crypto at the moment</p><p>I really want to offer another solution to another problem</p><p>But I just feel that the thing the crypto community needs most right now is simplicity</p><blockquote>“The only new invovation that will get in this eco-system is simplicity”</blockquote><p>Just to talk at eye level and present a problem &amp; solution in a simple way.</p><p>in crypto, simplicity is even more important because complexity has the tendency to centralize power in the hands of the specialized few, and that creates a lack of transparency which is exactly what caused the 2008 crisis where only a few specialized Finacial experts knew what was going on for better or for worst,<br>this very crisis is what decentralized currency systems aim to solve but can we solve complexity by replacing it with a different type of complexity?</p><p>So how do we make things a little simpler?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*QBJw-AxDfAY7plFG7zflJQ.png" /></figure><p>To be continued…</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=221676c697b0" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>