I work across backend systems, product infrastructure, and full-stack applications — with a particular interest in systems that have to handle real workflows rather than just demonstrate a technology.
Currently exploring the intersection of backend engineering and AI: retrieval systems, agent workflows, context management, and the infrastructure required to make AI applications reliable.
A personal AI system focused on giving an agent persistent knowledge and the ability to work with information beyond a single conversation.
I'm exploring the engineering behind:
- Retrieval and context management
- Document ingestion and indexing
- Long-term memory
- Agent workflows
- Tool execution
- Evaluation and reliability
Built around Python, FastAPI, PostgreSQL/pgvector, LangGraph and LLMs.
Building and experimenting with systems involving:
- Asynchronous and background processing
- Queue-driven workflows
- Concurrency control
- Data-intensive backend services
- API and database architecture
- Deployment and infrastructure
| Project | What it explores |
|---|---|
| Personal AI Agent | RAG, memory, retrieval, agent workflows and context management |
| Content Distribution Infrastructure | Distributed processing, queues, scheduling and concurrency |
| Time Tracker | Authentication, authorization, data modeling and backend architecture |
Systems Distributed systems · asynchronous processing · concurrency · backend architecture
AI RAG · retrieval · agents · memory · context engineering · LLM applications
Infrastructure Databases · Redis · Docker · AWS · deployment
Product Engineering APIs · full-stack systems · data modeling · reliability
Python · TypeScript · JavaScript
FastAPI · Node.js · Express · React · Next.js
PostgreSQL · MySQL · MongoDB · Redis
LangChain · RAG · Embeddings · LLMs
Docker · AWS · GitHub Actions