<?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 Monica Mohan on Medium]]></title>
        <description><![CDATA[Stories by Monica Mohan on Medium]]></description>
        <link>https://medium.com/@monicamohangis?source=rss-d83b0078a976------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*hbZV5Nysr6vFzcTXsKTDyw.jpeg</url>
            <title>Stories by Monica Mohan on Medium</title>
            <link>https://medium.com/@monicamohangis?source=rss-d83b0078a976------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 13:54:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@monicamohangis/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 Spatial Architect’s 2026 Roadmap]]></title>
            <link>https://monicamohangis.medium.com/the-spatial-architects-2026-roadmap-a0f2fea39bf1?source=rss-d83b0078a976------2</link>
            <guid isPermaLink="false">https://medium.com/p/a0f2fea39bf1</guid>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[spatial-analytics]]></category>
            <category><![CDATA[building-data-pipelines]]></category>
            <category><![CDATA[gis]]></category>
            <dc:creator><![CDATA[Monica Mohan]]></dc:creator>
            <pubDate>Sat, 07 Feb 2026 20:18:22 GMT</pubDate>
            <atom:updated>2026-02-07T20:18:22.702Z</atom:updated>
            <content:encoded><![CDATA[<p>In my last post, we explored how a <strong>Spatial Architect</strong> mindset can 10x your individual productivity. But in 2026, the game has evolved. We aren’t just writing better scripts anymore; we are building <strong>Agentic Systems</strong> — self-healing, autonomous workflows that don’t just execute tasks, but own outcomes.</p><p>If you’ve been following my <a href="https://github.com/gisbymonica/agenticforgis/tree/main/geoagent-starter"><em>GeoAgent-Starter repo</em></a>, you’ve seen the “Hello World” of this new era. Today, let’s refine the “how-to” and look at the definitive tech stack for getting started on your own Agentic AI journey.</p><h3>🏛️ The 2026 Spatial Architect Tech Stack</h3><p>The “Agentic Journey” requires a shift from linear chains to <strong>graph-based orchestration</strong>. Here is the lean, modern stack we’ve used in our repo:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k9YA_qoAccJ43CI8qeaE9A.png" /></figure><h3>The “Agentic Journey”: How to Get Started</h3><p>Transitioning from a “Analyst” to a “Spatial Architect” doesn’t happen overnight. Here is the 3-step path to building your first GeoAgent:</p><h4>1. Stop “Chaining,” Start “Graphing”</h4><p>Traditional GIS scripts are linear (A -&gt; B -&gt; C). If Step B fails (e.g., a CRS mismatch), the script breaks. In the Agentic Journey, we use <strong>LangGraph</strong>. This allows the agent to <strong>loop</strong>. If the agent detects a topology error, it doesn’t crash; it calls the repair_geometry tool and tries again.</p><h4>2. Empower the “Effectors” (Tools)</h4><p>An LLM is just a brain in a jar. To act, it needs <strong>Tools</strong>. In our starter repo, we built specialized tools like repair_and_reproject.</p><ul><li><strong>Rule of Thumb:</strong> If you find yourself doing a manual task in QGIS more than three times a day, turn that logic into a Python @tool.</li></ul><h4>3. Implement “Strict Spatial Governance”</h4><p>Agents can be over-confident. In our repo, we set the environment flag OGR_GEOMETRY_ACCEPT_UNCLOSED_RING=NO. This forces the agent to handle data corruption explicitly rather than letting GDAL &quot;silent-fix&quot; it. <strong>Trust, but verify</strong> is the Spatial Architect&#39;s mantra.</p><h3>Real-World Scenarios: From Scripts to Agents</h3><h4>Scenario A: The Self-Healing Urban Planner</h4><p><strong>The Problem:</strong> An analyst uploads a new city parcel dataset, but it has sliver polygons and unclosed rings that break the intersection tools. <br><strong>The Agentic Solution:</strong> Instead of failing, the agent:</p><ol><li>Detects the TopologyException.</li><li>Calls repair_geometry (using our make_valid() tool).</li><li>Re-runs the intersection and provides the final acreage.</li></ol><h3>Scenario B: Autonomous Logistics Routing</h3><p><strong>The Problem:</strong> A delivery route is planned, but a temporary road closure is detected in a live feed. <br><strong>The Agentic Solution:</strong> The agent uses a <strong>Mapbox MCP tool</strong> to fetch real-time traffic, identifies the blocked segments, and recalculates the isochrones — all while notifying the driver in natural language.</p><h3>🚀 Get the Starter Pack</h3><p>I’ve open-sourced the <a href="https://github.com/gisbymonica/agenticforgis/tree/main/geoagent-starter"><strong>GeoAgent-Starter</strong> </a>to give you a head start. It includes:</p><ul><li><strong>Self-Healing Geometry Logic:</strong> Auto-fix for unclosed rings and self-intersections.</li><li><strong>Streamlit Portal:</strong> A drag-and-drop map interface for interacting with your agent.</li><li><strong>3.14 Compatibility:</strong> Fully engineered for the latest Python runtimes.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qDPHy79VXTUaXN-m9D27IA.png" /></figure><p>The journey from GIS Professional to Spatial Architect is about building systems that “self-heal” , not the ones that break at the slightest inconvenience.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a0f2fea39bf1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[10x Your Spatial Productivity: The “Agentic” Secret No One Is Talking About]]></title>
            <link>https://monicamohangis.medium.com/10x-your-spatial-productivity-the-agentic-secret-no-one-is-talking-about-ae56897777d1?source=rss-d83b0078a976------2</link>
            <guid isPermaLink="false">https://medium.com/p/ae56897777d1</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[data-engineering]]></category>
            <category><![CDATA[gis]]></category>
            <category><![CDATA[geospatial-data]]></category>
            <dc:creator><![CDATA[Monica Mohan]]></dc:creator>
            <pubDate>Tue, 27 Jan 2026 18:03:25 GMT</pubDate>
            <atom:updated>2026-01-27T18:03:25.570Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s be honest: your current GIS ETL pipeline is a ticking time bomb.</p><p>You’ve probably spent weeks – if not months – perfecting Python scripts and FME workbenches. You’ve accounted for every null value and every geometry error you can think of. Yet, all it takes is one “creative” data entry from a field tech or a subtle schema change from a vendor to send your entire database into a tailspin.</p><p>In a world of real-time satellite feeds and IoT sensors, the “if-then-else” way of handling spatial data isn’t just slow – it’s becoming a liability.</p><h4><strong>The Maintenance Trap</strong></h4><p>We’ve all been there. You run a massive update on a Friday afternoon, and suddenly 15% of your polygons are floating in the middle of the ocean. Reason? Just a coordinate system mismatch that your script wasn’t programmed to catch.</p><p>Traditional ETL is <strong><em>deterministic</em></strong>. It’s a machine that does exactly what you tell it, even if what you told it to do is objectively wrong for the data you actually received.</p><p>We’re stuck in a cycle of:</p><ul><li>Fixing broken schemas instead of analyzing trends.</li><li>Writing endless Regex for address cleaning that still fails 20% of the time.</li><li>Manual QA that’s physically impossible to scale.</li></ul><p><strong>Why I’m Switching to “Agentic” GIS</strong></p><p>The buzzword “AI” is everywhere, but for those of us in the weeds with spatial data, the shift to Agentic Workflows is a genuine paradigm shift. We aren’t talking about a basic chatbot. We’re talking about autonomous agents that can actually reason through a spatial problem.</p><p>While your old script sees a syntax error and simply dies, an AI Agent sees a context clue. It recognizes that “Stn Rd” and “Station Road” are the same entity across two different datasets and merges them without you needing to write a single line of logic for it.</p><h4><strong>The 3-Step Reality Check</strong></h4><p>If you want to stop being a “Data Janitor” and start being a Spatial Architect, this is the path forward:</p><ol><li><strong>Stop Hard-Coding Everything:</strong> Start wrapping your GIS functions (clipping, buffering, re-projecting) as “Tools” that an LLM can call.</li><li><strong>Build a “Self-Healing” Loop:</strong> Use frameworks like LangGraph to let the agent “try again” if a join fails. If it hits a CRS error, it should be smart enough to look at the metadata and re-project on the fly.</li><li><strong>Use Multi-Agent Systems: </strong>Have one agent specialized in geometry validation and another in attribute normalization. Let them argue over the data quality before it ever hits your production database.</li></ol><h4><strong>The Bottom Line</strong></h4><p>The gap between the shops using static ETL and those using AI-augmented pipelines is widening fast. One group is drowning in technical debt; the other is actually getting work done.</p><p>If your GIS strategy still relies on rigid logic for a messy, non-rigid world, you aren’t just behind the curve – you’re off the map.</p><p>Most people think this tech is years away. It’s not. In the next part of this series, I’m sharing the framework I used to build my first autonomous GIS crew. Stay tuned for the roadmap to becoming a Spatial Architect.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ae56897777d1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Installing SSL Certificate for GeoServer/Jetty]]></title>
            <link>https://monicamohangis.medium.com/installing-ssl-certificate-for-geoserver-jetty-1d325e23e0e1?source=rss-d83b0078a976------2</link>
            <guid isPermaLink="false">https://medium.com/p/1d325e23e0e1</guid>
            <category><![CDATA[https]]></category>
            <category><![CDATA[jetty]]></category>
            <category><![CDATA[ssl]]></category>
            <category><![CDATA[geoserver]]></category>
            <category><![CDATA[security]]></category>
            <dc:creator><![CDATA[Monica Mohan]]></dc:creator>
            <pubDate>Thu, 17 Jun 2021 09:37:23 GMT</pubDate>
            <atom:updated>2022-12-14T08:58:50.204Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="GeoServer Logo" src="https://cdn-images-1.medium.com/max/200/0*5TcUhnu5yJSkp1w2.png" /><figcaption>GeoServer</figcaption></figure><p>Here’s a short tutorial on installing SSL Certificate for Jetty Server that comes bundled with GeoServer.</p><p>Disclaimer: The following steps are performed and tested in Windows OS</p><h3>Generate PFX Certificate</h3><p>You can create self-signed certificate and export it as a pfx file in Powershell as administrator. For this, run the following command:</p><blockquote>$cert = New-SelfSignedCertificate -certstorelocation cert:\localmachine\my -dnsname &lt;domain name&gt;</blockquote><blockquote>$pwd = ConvertTo-SecureString -String ‘password’ -Force -AsPlainText</blockquote><blockquote>$path = ‘cert:\localMachine\my\’ + $cert.thumbprint</blockquote><blockquote>Export-PfxCertificate -cert $path -FilePath m:\certs\powershellcert.pfx -Password $pwd</blockquote><p>Install the certificate by simply double-clicking the newly generated pfx file.</p><h3>Generate KEYSTORE file</h3><p>Keystore file is required by Jetty for SSL configuration. Hence, convert the PFX file to JKS Keystore and then install it on the Jetty Server that comes with GeoServer.</p><ul><li>Go to ‘bin’ folder of JDK and run the following command in cmd:</li></ul><blockquote>keytool -importkeystore -srckeystore m:\certs\powershellcert.pfx -srcstoretype pkcs12 -destkeystore m:\certs\keystore -deststoretype JKS</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a58FBu-AYJqAc-XgQRGvZA.png" /><figcaption>Import keystore from PFX certificate file</figcaption></figure><p>**Tip: Use the same password that you used to generate the pfx file</p><ul><li>Now, execute the following command to verify if the created keystore is a PrivateKeyEntry.</li></ul><blockquote>keytool -list -keystore m:\certs\keystore -storepass password</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bU2XgdFreM3byw9RBNt2UA.png" /><figcaption>List Keystore</figcaption></figure><ul><li>Copy the generated keystore to %GEOSERVER_HOME%\etc\keystore (make a backup of the existing keystore)</li></ul><h3>Configure Jetty Server</h3><ul><li>Open the start.ini in %GEOSERVER_HOME% and copy the following to after — module=http</li></ul><blockquote>#SSL</blockquote><blockquote>— module=ssl</blockquote><blockquote>jetty.ssl.port=8443</blockquote><blockquote>jetty.sslContext.keyStorePath=etc/keystore</blockquote><blockquote>jetty.sslContext.trustStorePath=etc/keystore</blockquote><blockquote>jetty.sslContext.keyStorePassword=password</blockquote><blockquote>jetty.sslContext.keyManagerPassword=password</blockquote><blockquote>jetty.sslContext.trustStorePassword=password</blockquote><blockquote>— module=https</blockquote><blockquote>jetty.httpConfig.securePort=8443</blockquote><ul><li>Download the jetty-distribution file (it must be the same version as in the geoserver — geoserver 2.21 uses jetty 9.4.48) from maven eclipse site and extract the contents.</li></ul><p><a href="https://repo1.maven.org/maven2/org/eclipse/jetty/jetty-distribution/9.4.48.v20220622/">Central Repository: org/eclipse/jetty/jetty-distribution/9.4.48.v20220622 (maven.org)</a></p><p>**Tip: You can cross check the jetty version in %GEOSERVER_HOME%/lib</p><p>(3 files are required — ssl.mod, jetty-ssl-context.xml and jetty-util-xx.jar)</p><ul><li>Copy ssl.mod from /modules to %GEOSERVER_HOME%\modules</li><li>Copy jetty-ssl-context.xml from /etc to %GEOSERVER_HOME%\etc</li><li>Copy jetty-util-x.x.xx.jar to any path on the machine and navigate to this path in cmd</li></ul><p>Run this command:</p><blockquote>java -cp jetty-util-&lt;JettyVersion&gt;.jar org.eclipse.jetty.util.security.Password password</blockquote><p>and obtain the obfuscated password (OBF:….)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YAtX0J82Y-yn9LDINnyP9g.png" /><figcaption>Generating obfuscated password</figcaption></figure><ul><li>Copy the OBF password and replace the existing OBF password in jetty-ssl-context.xml</li></ul><p>**Tip: If port needs to be changed, make changes in jetty-ssl.xml (secure.port) and in start.ini</p><ul><li>Restart the GeoServer service in Services.msc or Restart using start.bat in %GEOSERVER_HOME%</li></ul><p>Now, you will be able to access secure GeoServer using https://&lt;domain&gt;:&lt;port&gt;/geoserver/web</p><p><em>Cheers!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1d325e23e0e1" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>