NYC Taxi Ride Demand Prediction
Built a serverless Python/PySpark pipeline with Lambda, S3, Glue, Athena, and SQL to ingest, validate, and transform 10M+ trip records into curated datasets for forecasting and a Power BI dashboard.
Hi! I'm Vaibhav, a data platform engineer. I build reliable pipelines and platforms with Python and SQL on AWS and GCP, from data ingestion and validation to analytics and applied AI.
Built a serverless Python/PySpark pipeline with Lambda, S3, Glue, Athena, and SQL to ingest, validate, and transform 10M+ trip records into curated datasets for forecasting and a Power BI dashboard.
Fine-tuned LLaMA-3 8B Instruct with LoRA on approximately 4.3K curated diabetes Q&A pairs and built a RAG pipeline over a 2,032-page clinical guideline. Evaluated with BLEU, ROUGE-L, BERTScore, and BART-MNLI.
I work across the full stack of data and AI systems, from infrastructure to interfaces. Every project is built for production reliability and measurable impact.
Learn More About MeI'm always interested in challenging problems and impactful projects. Whether you're looking to hire or collaborate, let's connect.
Portfolio-aware, chat-only assistant
I'm OpenClaw, Vaibhav's portfolio assistant. I can help with his background, projects, skills, and recruiter-style questions, and I can still answer general questions as a chat-only assistant.
10:43 PM
Conversations may be logged for monitoring and security.
Contact