Work
  • Jan 2025 - Present
    PowerSchool
    Senior Software Engineer
    • Shipped PowerBuddy for Data Analysis (PBDA), an agentic AI analytics platform that lets non-technical educators extract insights from their data. Designed the agent harness — personas, memory, tool calling, and state management — to scale natively across diverse data domains.
    • Cut inference costs by 80% and latency by 55% through context engineering, streamlined agent communication, and state-aware tool design on a serverless AWS Lambda / DynamoDB stack.
    • Built autonomous AI agents for large-scale data migration, generating complex SQL across ~100 domains in minutes — cutting customer migration costs by over 95% and human validation effort by ~90%.
  • Jun 2024 - Sep 2024
    Coherent
    Machine Learning Intern
    • Built a GPT-4o based multimodal chatbot to automate optical and electrical simulations, achieving 90% error-free simulations.
  • Jun 2018 - Jun 2023
    Samsung Research
    Lead Machine Learning Engineer
    • Developed an on-device multi-frame CNN in TensorFlow for real-time low-light video restoration on the Galaxy S23 — 4x better temporal consistency, 6% higher SSIM, and 200x lower complexity than state-of-the-art baselines.
    • Curated a 50K-video training dataset with real noise and synthetic motion to reduce motion blur and enhance low-light detail.
    • Mentored a team of engineers to a 10% performance boost on the Galaxy S23 FE image-processing pipeline by optimizing CPU core utilization and reducing inter-thread dependencies.
  • Jul 2017 - Dec 2017
    Siemens
    Research Intern
    • Developed an algorithm to extract mathematical formulae and charts from technical PDFs with 76% accuracy.
    • Built a chart classifier using transfer learning on GoogLeNet, achieving 91% accuracy.
  • May 2017 - Jul 2017
    Samsung Research
    Software Intern
    • Developed a closed-domain QA system for home appliances using Dialogflow and a seq2seq LSTM.
  • May 2016 - Jun 2016
    DataPhi Labs
    Data Science Intern
    • Built a customer churn prediction engine using random forest, achieving an F-score of 0.76.
    • Developed a dynamic insights-mining wrapper for user-specific retention strategies.