Post Job Free
Sign in

Data Science for Reliable High-Stakes Systems

Location:
San Jose, CA
Posted:
November 18, 2026

Contact this candidate

Resume:

Samyuktha Saravanan

+1-971-***-**** California, USA – 95134

*************@*****.*** sam3092000.github.io linkedin.com/in/samyukthalsaravanan TECHNICAL SKILLS:

Tools: Python, SQL, R, SAP, Databricks, Snowflake, Tableau, Power BI, Excel, Pandas, NumPy, Scikit-learn, LangChain, Hugging Face Transformers, Git, Docker, Airflow, AWS, Spark ML / AI: Regression, Classification, Clustering, NLP, TimeSeries Forecasting, Anomaly Detection, LangChain, RAG, Agentic AI EDUCATIONAL QUALIFICATION:

Master of Science in Statistics, GPA: 3.89/4.00 Aug 2022 – Dec 2023 Arizona State University

Bachelor of Engineering in Electronics and Communication, GPA: 3.46/4.00 Aug 2018 – May 2022 Anna University

WORK EXPERIENCE:

Sr Data Scientist (Full-time) Honeywell Aero Oct 2025 – Present

• Drove product analytics strategy across 5+ operational surfaces, building KPI dashboards for inventory health, on-time delivery, and cost variances to translate open-ended questions into actionable insights.

• Architected foundational data infrastructure for a newly formed data team, integrating SAP and 4+ supply chain systems into automated ETL pipelines enabling self-serve dashboards used daily by planning and procurement.

• Designed and currently deploying a GenAI-powered pipeline to extract and structure incident datasets from internal documents, accelerating decision-making for the Director of Procurement and enabling 15+ non-technical users to run independent analyses.

• Partnered cross-functionally with planning, procurement, and logistics teams to deploy forecasting and anomaly detection models that flag potential stockouts and delivery risks before issues impact production

• Reduced 25+ hours/week of manual planner/buyer workload by building and deploying AI chat agents using Python and LangChain, enabling natural language queries on stock levels and lead times for 3k+ active users.

• Built a Retrieval-Augmented Generation (RAG) system over internal supply chain documentation, enabling planners to query policies, SOPs, and historical decisions in natural language with source-cited responses. Data Scientist I (Full-time) Honeywell Aero Jan 2024 – Sept 2025 Data Strategy (Intern) May 2023 – Dec 2023

• Identified a $575M untapped revenue opportunity by building ML models combining internal product data with 7 high- dimensionality macroeconomic and industrial signals into a clear growth strategy for senior leadership.

• Replaced 40 hours/week of manual reporting by building automated SQL pipelines and dynamic Power BI dashboards for strategic planning across Air Transport and Business Aviation, ensuring continuity through a platform migration.

• Led migration from a legacy analytics platform to a new data source, collaborating cross-functionally to resolve IT bottlenecks and deliver custom Python and Tableau solutions.

• Improved flight hour calculation accuracy by 14% through end-to-end analytics and designed experiments comparing model approaches, directly enhancing Honeywell’s Maintenance Service Plan (MSP) pricing precision and strengthening OEM/APU sales positioning.

Data Science Intern Zoho Corporation India Nov 2021 – Jun 2022

• Analyzed ~20GB user interaction data across Zoho’s SaaS platforms to uncover usage patterns, supporting product optimization for Zoho Analytics and related tools using data visualization.

• Built ML classification models in Python to predict churn risk for Zoho Books users, engineering features from multiple raw sources through custom ETL pipelines processing unstructured SaaS telemetry data. PROJECTS:

Predicting which Public Health Factors have the greatest Impact on Life Expectancy in Python Jan 2023 – May 2023

• Predicted life expectancy using regression with various regressors against different classifier algorithms in Python. Experienced in creating and evaluating multivariate linear regression models to determine the best fit. Comparison Analysis of Machine Learning Techniques to Predict the Winner of Reality TV Show Jan 2023-May 2023

• Collected, preprocessed, and analyzed data from seasons 11 to 25, with total of 422 contestants, on 'The Bachelor' to build Machine Learning models using classification algorithms like Logistic Regression, Random Forests, Bayesian Additive Regression Trees(BART) and Support Vector Machines, where we trained models with historical data, analyzed important features, and conducted sensitivity analysis to evaluate model performance in different scenarios.



Contact this candidate