building things in AI agents, analytics, infra, data systems.
previously worked on LLMs, RLHF, small models, and a lot of data pipelines.
- AI agents & orchestration
- Real-time analytics & data infra
- LLM pre-training, fine-tuning, RLHF, small models
- Distributed training & inference at scale
- Systems that just work
- Codex fork — multi-agent team orchestration in Rust for OpenAI's Codex CLI. PR #1
- Multi-agent collaborative workspace — humans and AI agents in shared rooms, real-time collaboration
- LLM research — process reward models, test-time compute scaling, and code agents for SWE-bench
- ClickHouse/clickhouse-kafka-connect — core contributor, built multitenancy support (dynamic database routing, topic-based table mapping, virtual topic state management). PRs & issues
- agents-ui/agents-kit — ready-made components for AI agents
- meow-ssh — SSH auth via browser passkeys (Touch ID / Face ID)
- clickhouse-kafka-connect — my fork with multitenancy features before upstream merge
- email: me@abhishek.it
- linkedin: in/abhishekit
- location: remote