Seven years on backend systems at Cafebazaar and Divar. Now building production AI that holds up under real load.
Six case studies on what the system looked like, what I built, and what turned out to be hard.
Independent
Independent engagements building production AI for B2B SaaS clients - RAG, agent platforms, multi-tenant retrieval. The caching-and-fallbacks work that decides whether an LLM feature holds up in front of a real customer.
Cafebazaar
Tech lead for the ad platform at Cafebazaar (Iran's largest Android store, 51M users). Owned a 10+ engineer team through a live microservices migration and the systems work that grew ad revenue 10× over two years.
Cafebazaar
Shipped 15+ high-reliability backend APIs at 51M-user scale, decomposed the legacy payment monolith into independently deployable services, and cut data loss on the stats aggregation service from ~15% to under 1%.
I do my best work on systems that have to keep running at 3 AM.
Usually the headline problem is not the problem. Latency, downtime, a team that has slowed down: the cause tends to sit one layer below, in service boundaries, retry storms, or deploy coupling. I would rather find that than ship around it.
I mentored seven junior engineers at Cafebazaar and five were promoted internally. I am not big on force-multiplier framing, but the most useful thing a senior engineer can do is leave a team better at the job than they found it.
Away from the terminal I cook a lot and chase cuisines I have not tried, walk every city I visit deliberately off the tourist map, and keep a collection of useless trivia. Ask me about the connection between clicker lighters and clocks.
Read the longer versionPython (Django, FastAPI), Go, JavaScript / Node.js, C++, gRPC, REST, async APIs
RAG over PDF, DOCX, plain text, Embeddings and vector stores (Pinecone, Elasticsearch), Citation-anchored retrieval, Agent design under predictability constraints, LLM API integration (OpenAI, Anthropic, open-source), AI-assisted SDLC (Claude Code, evals, agentic dev loops)
Docker, Kubernetes, GCP, AWS, CI/CD pipelines, Linux / shell
PostgreSQL, Redis (caching, pub/sub, dedup), Apache Kafka, RabbitMQ, MongoDB
Grafana, Prometheus, Sentry, Distributed tracing, Custom telemetry tooling
Microservices and service boundaries, System design at scale, Code review and mentorship, Agile / Scrum leadership, Unit and integration testing
Email is the fastest path, and the resume has the context. I also take on one or two consulting engagements at a time, in backend architecture and AI integration.