The latest articles on DEV Community by alilo113 (@alilo113). https://dev.to/alilo113 https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F1976787%2F5a32b3e1-7241-4dfb-9aa4-d2e30e3bfaa5.jpeg https://dev.to/alilo113 en alilo113 Sat, 10 Oct 2026 07:47:50 +0000 https://dev.to/alilo113/how-i-built-a-lightweight-local-first-vs-code-ai-assistant-with-ollama-3nlm https://dev.to/alilo113/how-i-built-a-lightweight-local-first-vs-code-ai-assistant-with-ollama-3nlm <h2> 🚀 The Motivation </h2> <p>Cloud AI coding assistants are convenient, but subscription fees add up and sending enterprise or personal codebases to external APIs isn't ideal for privacy. </p> <p>I wanted a simple, zero-cost, local-first alternative that integrates directly into VS Code and hooks straight into <strong>Ollama</strong>. So, I built <strong>Echo</strong>.</p> <h2> 🛠️ How It Works Under the Hood </h2> <p>To keep the MVP fast and reliable, the architecture is entirely decoupled from heavy frontend frameworks:</p> <ol> <li> <strong>Extension Host (TypeScript):</strong> Listens for commands and sends direct HTTP requests to the local Ollama instance running at <code>http://localhost:11434</code>.</li> <li> <strong>Inline Completion Provider (TypeScript):</strong> Hooks directly into VS Code's native inline completion API (<code>vscode.InlineCompletionItemProvider</code>) to serve ghost-text suggestions as you type.</li> <li> <strong>Distribution:</strong> Packaged directly as a <code>.vsix</code> binary via GitHub Releases for complete control.</li> </ol> <h2> ⚡ Current Challenges &amp; What's Next </h2> <ul> <li> <strong>Hardware Variability:</strong> Users running different local models need smooth onboarding. My next step is adding automatic model detection via Ollama's API on startup.</li> <li> <strong>Streamlined Workflow:</strong> Refining how code snippets and prompts are handled directly inside the editor workflow.</li> </ul> <h2> 📦 Try It Out </h2> <p>Echo is available as an open-source MVP release on GitHub:</p> <ul> <li> <strong>GitHub Release:</strong> <a href="https://github.com/alilo113/echo-releases/releases" rel="noopener noreferrer">Download Echo MVP</a> </li> <li> <strong>Requirements:</strong> VS Code + <a href="https://ollama.com/" rel="noopener noreferrer">Ollama</a> running locally.</li> </ul> <p>I'd love feedback from fellow developers building local toolchains!</p> vscode ai ollama javascript