Yonatan yonatangross

AI-native solo operator. I run the systems a team used to.

Agents, RAG, infrastructure, full-stack — the work that used to need a team, run solo.
Building yonyon.ai in public · creator of OrchestKit.

I run a one-person AI software studio out of Tel Aviv. The pitch is simple: the systems a startup used to staff a team for — multi-agent backends, RAG, infra, CI, full-stack product — I design, ship, and operate solo, end to end. yonyon.ai is where I do it in public.

OrchestKit — my flagship, the AI dev toolkit for Claude Code

Stop explaining your stack. Start shipping. OrchestKit encodes production patterns — backend, frontend, AI/LLM, security, DevOps, testing, product — so Claude Code already knows your architecture before you type a word.

Claude Code

/plugin marketplace add yonatangross/orchestkit
/plugin install ork

Cursor: Settings → marketplaces → yonatangross/orchestkit → enable ork → new chat.

Starter 12 (skills.sh, any agent) — not the whole catalog:

npx skills add yonatangross/orchestkit -s doctor -s setup -s explore -s implement -s verify -s review-pr -s commit -s expect -s assess -s brainstorm -s create-pr -s remember

The implement skill is implement, not ork-implement.

Coverage spans backend (FastAPI, PostgreSQL, REST/GraphQL), AI/LLM (LangGraph, RAG, evals, prompt engineering), frontend (React 19, Next.js, TypeScript), workflows (PR review, implementation, brainstorming), testing, DevOps (CI/CD, Docker, Terraform), product, and security.

Repo · Docs · vs Superpowers

What I build

Multi-agent AI platforms

Production LangGraph systems — supervisor/worker patterns, RAG over pgvector, real-time WebSocket streaming, and LLM pool failover across providers.

AI operations infrastructure

End-to-end platforms with custom MCP servers, WhatsApp + comms integration, AI command centers, automated content pipelines, and LLM evaluation harnesses.