// AI Engineer
Building intelligent
systems & experiences
I design and build AI-powered products, infrastructure, and developer tools — from research to production.
What I'm doing now
$ cat now.txt
- → Exploring Context Graphs
- → Building Agents and Skills
- → Exploring Harness Engineering
▊
Featured Work
Selected projects I've built or contributed to.
local-rag-system
A production-grade local RAG system
View project →mini-context-graph
A context graph skill that also leverages the llm-wiki package to build small scale growing context graph
View project →local-voice-agent
A real-time voice conversational AI assistant that listens, understands, thinks, and speaks. Built for Apple Silicon Macs with local-first processing.
View project →agentic-ai
A no-nonsense repo to help people jumpstart their Agentic AI journey.
View project →excalidraw-gen-skill
Skill to plugin with your agents for using the excalidraw-gen tool for generating excalidraw diagrams
View project →Recent Blogs
Thoughts on AI, engineering, and building products.
Context Compaction: The Essential Skill of Effective AI Agents
The Great Migration: Why AI Agents are Trading “Menus” for “Laptops”
Prompt Caching: Slashing Latency and Cost
The Book
Agentic AI Demystified: Master the Essentials
A simple-to-follow book to get started with your Agentic AI journey. Covers fundamentals like ReAct and CodeAct frameworks, advanced concepts like memory, tool calling, and evaluation, then hands-on building with Amazon Bedrock Agents, Langflow, LangGraph, and agent communication protocols like Anthropic's MCP and Google's A2A.
- ▹ReAct & CodeAct frameworks explained simply
- ▹Memory, tool calling, evaluation & agent design
- ▹Build with Bedrock Agents, Langflow & LangGraph
- ▹Deep dive into MCP and A2A protocols
- ▹Real-world case studies, code examples & diagrams