Agentic Code Generation
Multi-agent workflow that writes and reviews executable test scripts from written test cases.
AI Engineer at IBM’s Infrastructure AI CoE, working across NLP, large language models and generative AI. Six years shipping production AI — multi-agent orchestration, retrieval-augmented generation over real enterprise documents, and Python backends that hold at a thousand concurrent users. Four and a half years at Accenture before this.
Multi-agent workflow that writes and reviews executable test scripts from written test cases.
RAG pipeline that reads specification documents — tables and diagrams included — and writes test cases from them.
[DESCRIBE YOUR IBM WORK — your LinkedIn has no description for this role, and it is the first thing a reader looks for.]
Agentic code generation and document-grounded test case generation.
Automated audit reporting on Azure OpenAI and Azure AI Search.
Prompt-driven image generation for product packaging.
Open to senior AI and ML engineering work — agentic systems, retrieval pipelines, and the unglamorous reliability engineering that keeps them running. Happy to talk about any of the systems on this site.
[WHICH EMAIL DO YOU WANT PUBLIC?]