Backend and AI engineer focused on Go, distributed systems, and production LLM infrastructure. I build things that run in the real world — not just on localhost.
- 🔭 Currently building: AskMyDocs — a production-grade RAG system in Go with Kubernetes, Qdrant, and OpenTelemetry
- ⚙️ Exploring: AI engineering, vector search, LLM evaluation pipelines, and Go concurrency internals
- 🧠 Interested in: systems that have operational weight — databases, inference infra, distributed coordination
Languages
Backend & Infrastructure
AI / ML
Production-grade RAG system built in Go, deployed on Kubernetes (GCP).
- Hybrid retrieval: BM25 + vector search (Qdrant) + cross-encoder reranking
- OpenAI embeddings, PostgreSQL metadata store, Redis caching
- OpenTelemetry + Prometheus observability, Langfuse eval tracing
- Full Kubernetes deployment with Helm
In-memory vector database implemented from scratch in Go.
- Cosine similarity search with concurrent read/write safety
- WAL-based persistence for crash recovery
- HNSW indexing (in progress)
- Metadata filtering support
Log-Structured Merge-Tree storage engine in Go.
- In-memory AVL tree (memtable) with SSTable flushing
- Write-Ahead Log for durability
- RESTful API for CRUD operations
Redis-compatible key-value server built in Go.
- Full RESP protocol implementation
- Concurrent request handling with goroutines
- Core command support: GET, SET, DEL, EXPIRE, TTL
CLI-based real-time network analysis tool.
- Berkeley Packet Filter (BPF) integration
- Interactive terminal UI (gocui)
- Time-series packet visualization
I'm always open to interesting conversations — backend systems, AI infrastructure, or anything at the intersection of the two.