Aratrika Basu

AI Researcher

San Francisco, California, United States7 yrs 3 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

Expert in building production ML and Gen-AI systems.
Proven track record of delivering measurable business impact.
Strong mentor and community contributor in data science.

Stackforce AI infers this person is a Data Scientist specializing in Retail and Healthcare analytics with expertise in Machine Learning and AI systems.

Contact

basuaratrika1993@gmail.com LinkedIn

Skills

Core Skills

Machine LearningOptimizationNatural Language Processing (nlp)Statistical ModelingApplied StatisticsAcademic AdvisingLeadership DevelopmentTransformer ModelsDemand ForecastingCustomer SegmentationFeature Engineering

Other Skills

A/B TestingAI System DesignAgentic AIAmazon Web Services (AWS)AnacondaApplied NLPBERT (Language Model)CoachingCross-functional CollaborationData AnalysisEconometric ModelingEmbeddingEnd-to-End ML LifecycleFirebaseGenAI

About

I am a Data Scientist building production ML and Gen-AI systems at Walmart. My work spans real-time LLM systems, optimization, uncertainty-aware statistical modeling, and experimentation, delivering measurable business impact at scale. I enjoy taking problems from research and prototyping to deployment, working across ML, applied statistics, and distributed systems. Previously, I have worked on forecasting, customer segmentation, and supply-chain analytics in healthcare and retail. I hold a Master’s in Applied Data Science from USC and enjoy mentoring junior engineers and contributing to data science communities.

Experience

7 yrs 3 mos

Total Experience

2 yrs 5 mos

Average Tenure

2 yrs 11 mos

Current Experience

Walmart

2 roles

Data Scientist/Machine Learning Engineer

Jun 2023 – Present · 2 yrs 11 mos · Sunnyvale, CA

Built an ML-driven invitation optimization system reducing over-quota by ~76% and closed-survey clicks by ~74%, saving $500K–$600K annually. Owned the end-to-end ML lifecycle, including a custom Random Forest regressor model, Optuna-based metric optimization, A/B Testing, MLflow tracking, Kafka deployment, and drift monitoring, supported by daily-refresh data pipelines.
Built a real-time hybrid GenAI-powered response quality system leveraging Transformers for
initial screening and LLMs for secondary validation, to flag irrelevant, nonsensical, profane, blurred, and NSFW responses with low latency and high accuracy. Used Milvus-based semantic alignment to reduce false positives, improving supplier experience, trust, and downstream revenue quality.
Built a monthly consumer sentiment index linking survey responses to transactional spend
using Dynamic Factor Models (DFM) and Structural Equation Modeling (SEM). Identified statistically significant latent factors driving customer spending; index is used to track economic outlook and spending behavior.

Machine LearningOptimization

Data Science III Intern

May 2022 – Aug 2022 · 3 mos · Sunnyvale, California, United States

Worked with the Data Ventures track at Walmart as part of their Customer Perception team.
Performed in depth analysis of DCCS model already implemented in their system. Identified the weakness of the model and suggested improvements.
Implemented a new Embedding model that generates high quality embeddings combining textual & transactional customer data, it can replace the old DCCS model and can be used for additional downstream tasks across Data Ventures track.
Got the opportunity to present my work to the entire Customer Perception team, Product team and also Director of Data Science. The work was tagged as a valuable work that can be implemented across different work streams and was appreciated for the same.

Transformer ModelsMachine Learning

Usc women in engineering

Student Mentor

Sep 2021 – May 2023 · 1 yr 8 mos · Los Angeles, California, United States

I actively contributed to the USC Women in Engineering mentorship program, helping students thrive in their academic pursuits.
Advised students on navigating the graduate school experience, enhancing their academic and professional skills.
Conducted workshops to provide valuable insights and foster a supportive community among women in engineering.
Developed strong mentorship relationships that encouraged personal and academic growth.

Academic AdvisingLeadership Development

Quinnox, inc.

Data Scientist

Jun 2019 – Dec 2020 · 1 yr 6 mos · Bengaluru, Karnataka, India

Built a statistical estimation framework generating defective-product rate bounds across
products, aligned with uncertainty-aware demand and quality estimation techniques used in large-scale supply chain systems; enabling prediction of expected defective units for proactive inventory planning and loss mitigation.
Built a time-series forecasting solution for order volume prediction within an Order Management System, supporting demand and inventory planning across 10+ product lines. Modeled trend and seasonality and operationalized
forecasts for planning teams.
Led teams of 3-5 members, training and mentoring them towards achieving organizational goals along with fostering team relations by regularly interacting with extended teams and clients based out of several locations in Europe.

Statistical ModelingDemand Forecasting

Tata consultancy services

Data Scientist

Aug 2016 – Jun 2019 · 2 yrs 10 mos · Noida Area, India

Built a customer segmentation framework for a healthcare client using behavioral and transactional (RFM) features, dimensionality reduction, and K-Means clustering to identify distinct engagement segments and support targeted outreach and reporting.
Cleaned, validated, and reconciled production data; performed root cause analysis on data discrepancies and implemented preventive controls to reduce future errors and maintain system reliability.
Collaborated with clients and stakeholders to get clarity on the requirements leading to the excellent quality of deliverable without any post-production defect.

Customer SegmentationFeature Engineering