AI • Data • ML
I build data & ML systems that scale. Currently shipping backend services and production ML pipelines at scale.
Fine-tune an open-weights LLM into an empathetic best-friend assistant and show measurable gains on EQ-Bench 3. Train with multi-objective SFT using auxiliary emotion and support-strategy heads, then optionally run DPO for alignment. Provide quantitative improvements and brief qualitative insights
An intelligent chatbot that predicts diabetes risk based on user inputs and provides health interventions.
A machine learning pipeline for real-time data processing and anomaly detection.
Aires is an intelligent resume screening application that leverages OpenAI's GPT models to automate candidate evaluation and shortlisting.
Neural network–powered face detection, emotion classification, voice analysis, and engagement scoring — running 100% client-side via WebGL-accelerated TensorFlow.js.
Experienced in building GenAI systems, agentic chatbots, RAG pipelines, high-performance LLM workflows, and scalable data engineering platforms.
Built a personalized AI tutoring platform using LLMs, RAG, and an agentic chatbot with major latency and cost optimizations.
Designed enterprise data models and data marts enabling reliable analytics and scalable foundations for ML and GenAI workloads.
Built incremental ADF pipelines and optimized SQL logic, improving performance and reducing compute cost for enterprise reporting.