Siva Movva

About Siva Movva

I have over 13 years of experience in extracting actionable insights and business value out of data. I am currently a Staff Data Scientist at Intel, doing some really exciting work in using the humongous amounts of structured and unstructured data that a semiconductor factory generates to make the factory equipment and labor more productive. In the process I have built proficiency with data extraction (SQL), data wrangling (Pandas), statistical thinking, data based decision making, dashboarding, data product ideation, driving user adoption and more recently using the exciting new Large Language Model (LLM) technology to solve some last mile problems - related to unstructured text data. In the past I have also managed teams of highly accomplished engineers in a high paced, high stress technology development environment. I joined Intel after getting my PhD in Chemical Engineering.

Experience

1. Staff Data Scientist

   Intel

   Jul 2022 — Present · Hillsboro, OR, US

   Developed and deployed a ML enabled recommendation system: to optimize preventive maintenance schedules for expensive chip processing equipment. Identified $12 million in potential savings and realized $4 million savings to date- Tools: SQL (data extraction), Python (Pandas for data cleaning and transformation), Seaborn and Matplotlib (plot generation), Power BI (KPI and ROI dashboards)- Techniques: Statistical methods (IQR, Weibull fitting), Change Point Detection on Time Series data, Machine learning (DBSCAN via SciKit-Learn)• Leveraged LLM (GPT-4o) to derive insights from unstructured text data, improving root cause comprehension by 15% for sub-optimal maintenance cycles through prompt tuning on internal ‘work-order’ documents.• Engineered a method to stitch together and analyze multiple equipment log files, identifying ~$100 million in CAPEX savings through a 5% improvement in Equipment efficiency- Tools: Python, Seaborn, Plotly, SQL, PostgreSQL dBAAS• Reduced Equipment Downtime: Innovated a technique to quantify maintenance behavior differences between various Intel factories, avoiding over hours of expensive equipment downtime annually- Tools: Python (Pandas), PowerBI, and ML (Unsupervised Clustering)• Technical Communication: Presented regular updates to the executive leadership team on the data solutions being built for the factory and the ROI being generated from using these solutions

Education

• UC Berkeley School of Information

  Master's degree, Information and Data Science (MIDS)

• Anna University Chennai

  B.Tech, Polymer Engineering

  2001 — 2005

• The Ohio State University

  PhD, Chemical Engineering

  2005 — 2010

Skills

• Leading Cross Functional Teams
• Collaborative Problem Solving
• Research and Development (R&D)
• Design of Experiments
• Programming
• Object-Oriented Programming (Oop)
• Python
• Materials Science
• Characterization
• Materials
• Dry Etch
• Polymers
• Jmp
• Chemical Engineering
• Process Simulation
• Engineering
• Thin Films
• Nanoparticles
• R&D
• Matlab
• Process Engineering
• Heat Transfer
• Powder X-Ray Diffraction
• Labview
• Cross-Functional Team Leadership