Hello! I'm
MS in Data Science @ University at Buffalo.
2× intern across NLP and ML. I build things that turn messy data into decisions — pipelines, models, and dashboards that actually get used.
MS Graduate
Python · SQL · NLP · LLMs · AWS · Power BI.
Open to full-time roles in Data Science, AI/ML Engineering, Data Engineering and Analytics.
Data doesn't tell stories on its own — I build the systems that make it speak. From NLP pipelines that cut manual review effort by 40% to ML models analyzing 3.5M+ taxi trips in near real-time, I focus on work that actually ships and scales. I recently graduated with an MS in Data Science from the University at Buffalo (GPA: 3.56), with hands-on experience across Python, SQL, AWS, LLMs, and BI tools. Right now I'm looking for full-time opportunities where I can bring both technical depth and product thinking to the table.
End-to-end obituary publishing platform using LLMs and NLP to automate content validation. Reduced manual review effort by 30–40% across 100+ records. Deployed in a live pilot to 50+ stakeholders.
Open projectStructured 50K+ logistics, inventory & marketing records with SQL & Python. Designed 10+ KPIs (on-time delivery, inventory turnover, LTV) in an interactive Power BI dashboard with what-if simulations.
Open projectML pipeline analyzing 3.5M+ NYC taxi trips to forecast hourly demand. Automated ETL on AWS cut processing time 40%. Normalized schema + tuned SQL indexes improved query latency 25%.
Open projectLLM vs ML sentiment analysis pipeline on 50K+ text samples. Fine-tuned DeBERTa-v3-base achieving 93.5% accuracy, outperforming DistilBERT and XGBoost by 8%. MLflow tracking for reproducibility.
Open projectEnd-to-end ride-demand prediction pipeline on Snowflake with ETL and LightGBM forecasting. Interactive dashboards comparing actual vs predicted rides across key NYC zones. Optimized SQL for 25% faster queries.
Open projectAI-powered music recommendation app that analyzes images to detect mood and suggests matching tracks. Combines computer vision with music discovery for a unique user experience.
Open projectApplied NLP techniques for text validation, content structuring, and workflow optimization in product features.
Implemented LLM-driven pipelines including prompt design and evaluation to enhance model reliability.
Conducted Python-based data analysis and supported backend/API integrations for end-to-end functionality.
Built ML solutions for credit card approvals, cancer detection & Uber analytics achieving 90–95% accuracy.
Deployed Telegram + LLM chatbots (OpenAI/Gemini) automating insights and cutting manual analysis time 35%.
Built web-scraping + NLP pipeline collecting 10K records to streamline text analysis and reporting.
sahilsubhasbhaivachhani@gmail.com · Buffalo, New York