Open to opportunities

Anupama Jha

@anupamajha

Message

Principal Data Scientist and ML leader, I build production AI systems that drive measurable business impact.

India

Message

What I'm looking for

I’m looking for a role where I can lead production ML/LLM systems end-to-end—partnering with product and engineering—to run smart experimentation, align stakeholders, and turn ambiguous problems into scalable, measurable data products.

I’m a Principal Data Scientist with 10+ years of experience leading high-impact machine learning systems across food-tech, telecom, healthcare, and ed-tech. I focus on production ML architecture, experimentation strategy, and turning ambiguous business problems into scalable data products.

Most recently at RebelFoods, I led 2–3 data scientists and delivered multiple concurrent ML initiatives, including an ETA prediction platform (p95 < 1.5s) that drove an 18% accuracy gain and reduced customer support volume by 5%. I also built an automated BERTopic pipeline that eliminated 30+ hours/month of manual reporting, and architected SKU-level inventory forecasting across 400+ locations, cutting food wastage by ₹10L/month (externally validated by AT Kearney).

Experience

Work history, roles, and key accomplishments

Principal Data Scientist

Rebel Foods

Jul 2021 - Oct 2025 (4 years 3 months)

Led a team of 2–3 data scientists to deliver production machine learning across ETA, forecasting, and NLP initiatives, improving model quality and adoption across Product, Operations, and Engineering. Built an ETA prediction platform achieving p95 < 1.5s, delivering 18% accuracy gains and reducing customer support volume by 5%; also built BERTopic customer-feedback analytics and an inventory forec

XGBoost ETA Accuracy AWS ECS Redis Feature Engineering BERTopic Demand Forecasting Stakeholder Management

Data Science Manager

Reliance Jio

Jul 2019 - Jul 2021 (2 years)

Contributed to a centralized data science function by designing real-time LSTM-based anomaly detection for cybersecurity use cases and enabling threat-response workflows. Fine-tuned BERT transformer models to provide high-velocity CEO sentiment intelligence, partnering with Engineering, Security, and Product to productionize ML systems.

Anomaly Detection BERT Fine Tuning NLP

Data Scientist (EdTech)

Barnes & Noble

Feb 2018 - Jun 2019 (1 year 4 months)

Built machine learning models to identify students at risk of poor academic outcomes (D/F/W) using static and dynamic engagement and performance features. Developed cohort-specific GLM models and worked with academic advisors and product teams to deliver interpretable, actionable insights for non-technical users.

Risk Modeling GLM Feature Engineering data preprocessing Cohort Analysis Model Interpretability

Associate Data Scientist

UnitedHealth Group

Aug 2015 - Jan 2018 (2 years 5 months)

Designed and implemented fraud detection machine learning solutions for insurance and provider fraud across highly imbalanced datasets. Achieved strong precision on the top 10% of ranked cases, contributing to nearly 60% of identified fraud and improving detection while reducing financial leakage.

Fraud Detection Imbalanced Learning Feature Engineering

Associate Software Engineer

NCR Corporation

Aug 2013 - Jul 2015 (1 year 11 months)

Developed SQL queries for data processing using joins, aggregations, and data merging. Conducted regular production system health checks and supported change implementations to maintain system stability.

SQL Data Aggregation SQL Joins Change Management Data Processing