Loading…
Projects, notes, articles and skills in one place. Tip: press Ctrl/⌘ + K from any page.
Home
Build. Learn. Ship. Repeat.
About
Who I am, my journey and what I am building now.
Skills
Technologies grouped by category.
Projects
Everything I have built, filterable by category.
Resources
Notes, cheat sheets and references.
Blog
Technical articles and build-in-public posts.
Experience
Where I build and where I trained.
Contact
Get in touch or send a message.
JEV Model Router
Local proxy for Claude Code that asks TypeSafe's Jev which model — Haiku, Sonnet or Opus — fits each new turn.
OpenAgent Eval
Local-first evaluation framework for RAG systems and AI agents — retrieval, generation and cost metrics behind one CLI and SDK.
ModelDock
Python-first model manager for local LLMs — discover, install, update and verify models across local runtimes from one SDK and CLI.
RepoHeart
Event-driven multi-agent system that runs inside GitHub Actions to triage issues, review PRs, repair CI and resolve conflicts.
run-git
Git automation CLI — add, commit, pull and push in one command, with smart commit messages, a TUI and one-step GitHub releases.
LinkedIn AI Comment Copilot
Chrome extension that drafts LinkedIn comments with a LangGraph pipeline behind a FastAPI backend, built to survive LinkedIn's dynamic feed.
RAGNOVA
End-to-end RAG chatbot with LangChain and Sentence Transformers for semantic search and LLM-powered Q&A.
mcp-web-search
MCP server for real-time web search, built on FastAPI with JSON-RPC and Server-Sent Events for agent integration.
AI Commit
Generates Git commit messages with local Ollama models — free, private and offline.
AutoML Studio
End-to-end ML platform that automates EDA, preprocessing, training and evaluation without writing code.
ML Mastery
Structured machine learning repository with hands-on notebooks, practical examples and quick-reference guides.
MCP (Model Context Protocol) Overview
What MCP actually standardizes, how the client/server/host roles fit together, and why it matters for connecting LLM agents to tools and data sources.
LangGraph Quick Reference
The core concepts and control-flow patterns in LangGraph — nodes, edges, state, and the conditional routing that makes it different from a plain LangChain chain.
RAG Evaluation Basics
A short reference on how to actually measure whether a retrieval-augmented pipeline is working — retrieval metrics, generation metrics, and the failure modes each one catches.
Claude Code Source Map Leak (March 31, 2026): A Deep-Dive Into What Happened, What Was Exposed, and What It Means
How a 59.8 MB source map file accidentally exposed 512,000 lines of Anthropic's proprietary Claude Code source code to the entire npm ecosystem—and what it reveals about AI tool architecture
Building AI-Powered CLI Tools with Local LLMs
How I built a privacy-first Git commit message generator using local large language models
LLMs
AI / GenAI
RAG
AI / GenAI
Agents
AI / GenAI
LangChain
AI / GenAI
LangGraph
AI / GenAI
MCP
AI / GenAI
RAG Evaluation
AI / GenAI
Sentence Transformers
AI / GenAI
Scikit-learn
AI / GenAI
Groq
AI / GenAI
Gemini
AI / GenAI
Ollama
AI / GenAI
FastAPI
Backend
REST APIs
Backend
Server-Sent Events
Backend
Celery
Backend
Streamlit
Backend
Next.js
Frontend
React
Frontend
Tailwind CSS
Frontend
Chrome Extensions
Frontend
Docker
Cloud & DevOps
GitHub Actions
Cloud & DevOps
Vercel
Cloud & DevOps
CI/CD
Cloud & DevOps
FAISS
Databases
ChromaDB
Databases
Redis
Databases
SQL
Databases
Git
Developer Tools
GitHub
Developer Tools
VS Code
Developer Tools
Claude Code
Developer Tools
PyPI packaging
Developer Tools
Python
Languages
C++
Languages
TypeScript
Languages
JavaScript
Languages