Jan 2024 – May 2024 · 4 mos · Palo Alto, California, United States · On-site

• Augmented the battery level survival analyses for model 3 and model Y
• Delivered multiple large scale fleet analyses (1B+ rows) using PySpark and Caspian DB for the fastest growing Energy segment leading to more effective warranty decisions.
• Expedited the Reliability testing for Cybertruck by creating interactive dashboards with Streamlit for tracking the JIRA tickets progress.
• Configured weekly Airflow dags for climate labelling for all of the Energy products - Megapacks, Superchargers, PowerWalls - to assess the impact of weather on product life.
• Created templates for more efficient analyses of fleet data thereby enabling the serving of analyses like Rainflow algorithm, Time spent at a value, etc much faster.
• Simplified a compliance procedure with NHTSA to ensure that the complex information scattered across multiple pdf files in organised into tables and viewed through a Tableau dashboard
• Improved the Reliability practices by upgrading the internal tools after detailed discussions with Reliability engineers.

Aug 2023 – Dec 2023 · 4 mos · Tempe, Arizona, United States · On-site

• Worked with Prof. Violet Syrotiuk for the course Advanced Computer Networks

Jun 2023 – Aug 2023 · 2 mos

• Contributed to the proprietary Data Science library in agile set up to automate the detection of new Limited Time Offers, Level Shifts and Change Points in historical data, Holiday Spikes and Dips, Stockout conditions etc.

Jan 2023 – May 2023 · 4 mos · Tempe, Arizona, United States · On-site

• Worked with Dr. Runger on a project funded by Intel.
• ◦ Fine-tuned the open-source Large Language Model Databricks Dolly v2 to generate an extractive summary about specific companies from the meeting notes and deployed it on AWS.
• ◦ Built an end-to-end pipeline in Python that ingests meeting notes and generates insights such as frequency of mentioned companies, sentiment about and extractive summary for each mentioned company.
• ◦ Used a transformer-based model to locate the position of numerical data in textual data and two characteristics of it - one, whether it has increased or decreased, and two, whether it is a percentage, dollar value, or general number.
• ◦ Used REBEL model to extract the relationship between entities linked to the numerical data from above experiment.

Aug 2022 – Dec 2022 · 4 mos · Tempe, Arizona, United States

• CSE 205 - Object Oriented Programming and Data Structures.
• Held weekly meetings with students and addressed their doubts regarding the programming assignments in Java and GUI assignments in JavaFX.

Jul 2021 – Jun 2022 · 11 mos · Bangalore Urban, Karnataka, India

• ◦ Reduced value leakage in supply chain of a beverage giant by 5% with accurate forecasts of demand using time series algorithms like ARIMA, TBATS, Croston and libraries like Prophet, Greykite.
• ◦ Expedited client onboarding from a week to 2 days by creating Python scripts that perform driver-based recursive forecasting using adaptable functions for data preprocessing, modelling, and generating explainable metrics.

Jul 2019 – Jul 2021 · 2 yrs · Bengaluru Area, India

• ● Collected Migration performance statistics to allow autonomous decisions for replication of data
• ● Enabled easy debugging for engineers by developing queries that output internal system variables
• ● Designed, Developed and documented the backend for Data Analytics and ML webapp in Python
• ● Improved performance of the webapp by SQL Query Tuning and Database (DB) optimisation
• ● Optimized Resource Commissioning by building dashboards that monitor DB size, app performance.
• ● Augmented the Extract-Transform-Load data pipelines by adding more data sources

May 2018 – Jul 2018 · 2 mos · Bengaluru Area, India

• ● Quantified key performance indicators (KPI) with data from multiple sources using complex SQL
• ● Delivered an interactive gamified leader-board using Tableau to track KPIs for different teams