At CareValue Health Limited, I design and deploy AI agents for WhatsApp and Telegram that provide conversational support and automate healthcare workflows. I’ve integrated LLM chat systems with contextual memory and multi-turn reasoning, and built full-stack applications with sub-100ms API response times.
At Turing, I trained and evaluated LLMs using RLHF and Side-by-Side evaluations, improving model accuracy by about 85%. I also built React and Next.js evaluation dashboards adopted by three internal teams, and contributed to AI benchmarking and multimodal data annotation.
