ex-Palantir · ex-Bloomberg · ex-Microsoft · London, UK
About
I design, build and ship production systems end to end — from infrastructure to the features people use. Over 10+ years I've worked as a founding CTO and as an engineer at Palantir, Bloomberg and Microsoft Research.
My focus is applied AI and NLP: LLM pipelines, retrieval-augmented generation, context engineering, knowledge graphs and semantic search on scalable serverless infrastructure. I write production code across the full stack every day, and I've repeatedly taken products from idea to live in front of enterprise customers.
Work Experience
Founding CTO and primary engineer of a VC-backed AI startup ($800k raised, Google for Startups Accelerator). I architected and wrote the production backend, Python LLM microservices and Next.js frontend across four products, while leading a team of 10.
- RFQ-automation and spend-analysis platform: a self-learning LLM system that turns unstructured procurement data into structured records and cost estimates. Up to 35% cost and 85% time savings for enterprise clients.
- LLM email copilot for SME order processing: groups threads, maps them to orders, tracks updates and auto-replies. Cut manual handling by ~70%.
- Enterprise knowledge base combining LLMs and knowledge graphs for organisation-wide semantic search, built for PE firms and enterprises.
- Financial-research engine over SEC filings using vector search and knowledge graphs, for hedge-fund analysts.
Skills
Speaks Turkish (native) · English (fluent) · German (B1–B2)
Get in touch
The best way to reach me is by email.