I work on end-to-end data science problems, from analyzing data and building machine learning models to deploying them in the cloud.
I have taught yoga to 100+ students and mentored junior data scientists.
Recognized by RIT for 2025–26 service through AI research outreach in the Rochester community, weekly yoga classes, fundraising, and helping students find their footing on campus.
Gorbel Inc.
New York · Jan 2025 – Jan 2026
I built an end-to-end ML decision-support system for operations and procurement, covering 10K–15K monthly transfers and an estimated 7% cost savings.
Rochester Institute of Technology
Rochester, New York · 2024 – 2026
Published two research papers at ASME FEDSM 2026 on using machine learning for heat sink cooling predictions.
Tata Consultancy Services
Bangalore, India · Aug 2017 – Jun 2023
Built and ran the data systems enterprise clients depend on: pipelines, services, and the batch jobs behind them, handed over in a state the next engineer could maintain.
A regression pipeline that predicts thermal resistance and pressure drop for heat sink geometries, trained on simulation data. Designers get an answer in milliseconds instead of waiting hours for a CFD run, which turns thermal design into something you can iterate on. Includes a feature-importance view so the model's reasoning stays inspectable.
Retail customers do not announce that they have left; someone who churned and someone simply due for their next shop look identical in the data. This project uses a campaign’s randomised holdout to measure what an offer actually caused, then asks which customers are worth spending on.
At a 5% budget, targeting by likely response returns 147 extra purchases per 1,000 customers against 32 for targeting by churn risk — but the keenest responders spend the least. Weight response by spend and the ranking inverts: ₽1.24M incremental revenue per 1,000 against ₽0.27M. Neither “who is leaving” nor “who responds” is enough on its own.
A pricing decision product rather than a notebook: a pricing user enters vehicle details and the system predicts market value, recommends an acquisition offer, estimates margin after reconditioning, flags quotes needing approval, and records what the human did with the recommendation.
A single view over ingestion jobs, freshness, and cost across a Databricks lakehouse — so the team stops finding out about a broken pipeline from the people downstream.
Models trained on parametric simulation data estimate thermal resistance across fin geometries and flow conditions, with an error analysis showing where the surrogate stays reliable and where it should defer back to full simulation.
The tools I reach for most days.
The work I do around campus, outside the code.
Volunteer instructor leading weekly power yoga sessions for students, with fundraising classes open to the wider campus.
Ongoing member, with Level 1: Presentation Mastery completed. The regular speaking practice shows up directly in how I present research and run design reviews.
Supporting international student life through orientation help, campus events, and the day-to-day work of helping people settle into a new place. Recognised with the Outstanding International Student Service Award.
Recommendations from managers and colleagues at Tata Consultancy Services, RIT, and my co-op team.
Over seven months she took complete ownership of designing and implementing a modern data and ML engineering pipeline. She built the ingestion workflows in Azure Data Factory, transformed the data through Azure Data Lake and Databricks, and deployed models through Azure ML Studio with full MLOps integration. She stepped into a team leadership role naturally and delivered with minimal guidance.
Reetu developed predictive models for inventory demand and stock optimization during her data science co-op. She turned complex operational data into decisions people could act on, and she communicated the results clearly to technical and non-technical stakeholders alike.
Reetu is a Python developer with a good understanding of the language and its frameworks. Her ability to write clean, efficient, structured code is commendable, and her aptitude for problem solving is one of her greatest strengths.
I worked with Reetu on DevOps projects and she consistently impressed me with her deep understanding of CI/CD, automation, and cloud infrastructure.
Reetu is energetic, curious, and keen to learn. Having worked with her for over a year, I can say she puts her best into everything, and her ideas carried real weight in team discussions.
No agenda required. Bring a role you're hiring for, a problem you're stuck on, or a paper you'd like to talk through.
Software engineering, machine learning, and data roles. If you're building a team that does the work carefully, I'd like to hear about it.
Gorbel
RIT
TCS