Code Runtime and iPaaS for AI Agent — execute Bun/Python snippets at scale via POST API + integrate with 800+ tools (N8N for AI Agents)

An example application that exposes a multi-step, quality-controlled LLM workflow built with LangGraph through a FastAPI service.

A self-hosted, dockerized agentic coder with Kilocode and Codex support

Kubernetes operator that manages AI agents as first-class resources

A repository-based workflow for coordinating AI development tools using deterministic, snapshot-driven shared context.

Agentic flow for scraping data using SearXNG.

Multiple specialized agents collaborate to create content. Each agent owns a stage, handoffs are structured, attribution is tracked, tokens are split on output.

Production-grade agentic research system with schema-enforced verifiable citations to eliminate AI hallucination. Built with LangGraph & FastMCP.

MedGemma Clinical Robustness Assistant Multi-agent dermatology diagnosis system with safety-governed agentic pause and evidence-grounded RAG (Kaggle Med-Gemma Impact Challenge).

Langgraph-powered agent that runs a coding task in gVisor

Senior AI Engineer @ Coinbase | Architecting Enterprise LangGraph & RAG Systems | Bridging Philosophy & Distributed Systems

An append-only ledger convention (HITL.md) to close the loop with AI coding agents by logging human rulings on taste, scope, and when to ask.

Production-grade Agentic RAG system for luxury chauffeur dispatch. Built with Azure OpenAI, Semantic Kernel, and atomic SQL booking logic to handle real-time fleet orchestration and long-term customer memory.

AI-driven career path optimization and skill-mapping utilizing LLMs, RAG architecture, and Vector Embeddings

A free, comprehensive video course teaching developers to build production-ready AI agents from scratch. Covers Python, LLMs, RAG, multi-agent systems, security, and cloud deployment across 15 modules with 100+ videos and 25+ hands-on projects. Built as a React-based static site hosted on GitHub Pages.

Local development CLI for running AI agent pipelines without a Kubernetes cluster

LangGraph implementation of ReWOO: Reasoning Without Observation framework

Helm charts for kubeswarm - AI agents as Kubernetes-native resources

AI-powered Multi-Agent Lead Intelligence Platform for utility cost-reduction consulting. Uses specialized agents (Scout, Analyst, Writer) to discover high-spend companies, enrich and score them as leads, and automate personalized outreach with human approval at every step

A terminal launcher for agent CLIs — Claude Code, Codex CLI, OpenCode, Gemini CLI, DeepSeek-TUI — that pairs each session with a self-contained template (mission, playbook, rules, tools, subagents, persona) and clones it into a fresh isolated chat every time you start working on something new.