Developer Relations & Developer Experience leader. I build developer ecosystems around AI agents, and I ship one in public every day.
Currently consulting with EVM Systems and quirq.ai on developer experience and agent tooling. Previously Developer Platform Lead at GaiaNet, and Platforms & Community at AngelHack and Tribe.
275+ public AI builds since Jan 1, 2026 | 95,000+ developers reached | agent-office at 327★
- Consulting with EVM Systems and quirq.ai on DevEx and agent tooling
- Shipping a public AI build every day at dailybuild.xyz, 275 so far, each with a repo, a write-up and a demo
- Writing deep dives on agent infrastructure on dev.to: capability sandboxes, LLM judges, time-travel debuggers, memory consolidation
- Open to consulting, advisory and speaking engagements in AI agents and developer experience
agent-office | TypeScript | 327★ | 104 forks
Watch AI agents walk to desks, think, collaborate, hire interns, assign each other tasks, run code and search the web, all rendered in real time as pixel art with persistent state. A working reference for multi-agent orchestration you can watch rather than a diagram.
dailybuild.xyz | 275+ builds since Jan 1, 2026
One shipped AI project every day, in public. Everything from an LLM council and a six-model Rubik's Cube decision arena to an offline medical scribe and a carbon-credit agent market.
clonewriter | TypeScript | 11★
A 100% private AI agent that writes in your personal style, running entirely on local LLMs via Ollama. Nothing leaves the machine.
- Built 40+ integration demos for autonomous AI agents (dev.to/gaiaai)
- Shipped open-source repos including Gaia Cookbook
- Ran workshops, talks and community events:
- OpenSearchCon 2025 India
- Agentic Ethereum Hackathon by ETHGlobal
- MotherAI
- ETHGlobal Singapore 2024
- OpenSource India 2024 and 2025
- Managed multi-country developer ecosystems across India, Sri Lanka, Nepal and Maldives (95K+ developers)
- Built an ambassador program with 400+ ambassadors a year
- Ran 10+ hackathons a year plus innovation programs for AWS, Shopify and IBM
- Product Manager at iBot (Nestlé, HUL, Elkay IoT dashboards)
- Head of Technology at Deep Red Ink (SaaS, analytics, APIs)
- MCP Dev Summit Mumbai 2026
- Open Source India 2024 | 2025
- OpenSearch India 2025
- ETHGlobal / ETHDenver 2025
- Devcon Bangkok 2024
- Google Cloud Community Days 2025
- Hacker houses, meetups and workshops across APAC
Core
- AI agents & LLM integrations
- Developer experience (DX) & tooling
- Product strategy & platform engineering
- Community architecture & ecosystem growth
Stack
Agent frameworks (Burr, Agno, Strands) · local LLMs (Ollama, LM Studio) · MCP · evals and LLM-as-judge · SDK and API design · developer programs
- I built the smallest version of local semantic search for macOS
- Two local decision models play chess, and every probability is on screen
- How I made a small decision model's belief visible, and what that taught me about building honest AI demos.
- Your agent says it addressed the review comments. Here's how you find out if it actually did.
- Four AI agents argue over one Penpot canvas until the design is good
