Backend/infrastructure engineer with ~1 year of production experience. Previously MTS at Trelium (YC F25), building sandbox and workflow infrastructure for AI agents.
Production work:
- Reduced persistent workspace setup from ~70s to ~10s and sandbox recovery from 486s to 14s, preserving sessions across restarts.
- Made workflow registration idempotent across retries, duplicate hooks and out-of-order deployments, cutting failures from ~90% to ~1%.
- Built reconnectable multi-client agent streaming with one canonical execution per task.
- Previously the first engineer at Muze, where I built the entire MVP solo in 30 days, including a Go backend handling 10k+ concurrent WebSocket connections at sub-100ms latency.
Systems projects:
- lsm-tree-go: Go storage engine with mmap WAL, bloom filters and compaction. Profiled to ~2M writes/sec and 5x BoltDB throughput on random writes.
- vase: Lock-free Go pub/sub broker with mmap WAL. ~62x Redis throughput at 100 subscribers in my benchmark.
- Engram Engine: Durable agent-memory layer using event sourcing and WASM isolation, with sub-1ms sandbox overhead.
Looking for Backend Engineer, Full-Stack Engineer, Founding Engineer, Infrastructure, Platform or Distributed Systems roles. Open to startups and established companies, remote or onsite, in India or internationally.
Remote: Yes (worldwide)
Willing to relocate: Yes, including internationally with visa sponsorship
Technologies: Go, Node.js, TypeScript, JavaScript, React, Next.js, distributed systems, Linux, PostgreSQL, Redis, Kafka, Docker, gRPC, TCP/WebSockets, concurrency, mmap, io_uring, storage engines
Résumé/CV: https://drive.google.com/file/d/11ASup-0rcQivWNIMR6mOoYqEkhe...
Email: [email protected]
GitHub: https://github.com/AasheeshLikePanner
Backend/infrastructure engineer with ~1 year of production experience. Previously MTS at Trelium (YC F25), building sandbox and workflow infrastructure for AI agents.
Production work:
- Reduced persistent workspace setup from ~70s to ~10s and sandbox recovery from 486s to 14s, preserving sessions across restarts.
- Made workflow registration idempotent across retries, duplicate hooks and out-of-order deployments, cutting failures from ~90% to ~1%.
- Built reconnectable multi-client agent streaming with one canonical execution per task.
- Previously the first engineer at Muze, where I built the entire MVP solo in 30 days, including a Go backend handling 10k+ concurrent WebSocket connections at sub-100ms latency.
Systems projects:
- lsm-tree-go: Go storage engine with mmap WAL, bloom filters and compaction. Profiled to ~2M writes/sec and 5x BoltDB throughput on random writes.
- vase: Lock-free Go pub/sub broker with mmap WAL. ~62x Redis throughput at 100 subscribers in my benchmark. - Engram Engine: Durable agent-memory layer using event sourcing and WASM isolation, with sub-1ms sandbox overhead.
Looking for Backend Engineer, Full-Stack Engineer, Founding Engineer, Infrastructure, Platform or Distributed Systems roles. Open to startups and established companies, remote or onsite, in India or internationally.
Writing: https://medium.com/@ashishrathour1102