At Thoughtworks, I architect production AI-agent services with tool calling, retrieval, routing, and human approval. I also designed LLM API integrations and RAG pipelines using Pinecone, Weaviate, and FAISS.
I shipped interfaces for streaming agent conversations and workflow reviews, and designed services and event-driven processing for long-running AI operations. I mentor engineers through code reviews and technical guidance on AI-agent and LLM development.
At Cera, I productionized LLM and ML features for healthcare workflows. I partnered with data scientists to move experimental models into production services, including inference APIs, deployment, and reliability.
At Seargin, I trained and integrated ML models for customer-facing products and designed full-stack products and microservices. I also deployed containerized services on AWS and improved quality and performance through testing and optimization.
