# Vaishnav Potlapalli > ML Engineer Location: San Francisco, California, United States Profile: https://flows.cv/vaishnavpotlapalli ## Work Experience ### Software Engineer - AI/ML @ Sedai Jan 2025 – Present | San Francisco, California, United States Worked on building a platform for large-scale autonomous cloud optimization across Databricks, Snowflake & EMR using heuristic algorithms, time-series forecasting. Currently building Sedai’s foundational GenAI offerings. ### Founding Software Engineer , Machine Learning @ Floma Jan 2025 – Jan 2025 Building LLM based Agents for marketing ### Graduate Teaching Assistant @ NYU Center for Data Science Jan 2024 – Jan 2024 DS-GA.1011 Natural Language Processing with Representation Learning, Fall 2024 ### ML Intern @ Floma Jan 2024 – Jan 2024 ### Research Assistant @ MBZUAI (Mohamed bin Zayed University of Artificial Intelligence) Jan 2022 – Jan 2023 | Abu Dhabi Emirate, United Arab Emirates Developed and implemented PromptIR, a novel visual transformer-based prompt-learning framework for comprehensive image restoration and enhancement. This framework outperformed existing methods, improving by 0.9 dB across dehazing, deraining, and denoising benchmarks and was featured at NeurIPS 2023. Additionally, I adapted continual learning techniques like L2P and DualPrompt for video action recognition, boosting accuracy by over 10% and BWF by 14% on various benchmarks. ### Machine Learning Engineer @ Dhan AI Jan 2020 – Jan 2022 | Hyderabad, Telangana, India Enhanced the NER engine by developing a BERT-based classifier ensemble, achieving a accuracy improvement in entity recognition and sentiment classification, significantly boosting our primary product, a Patient Life Cycle Management Chatbot. Overhauled the application testing pipeline, leveraging increased parallelism and Redis caching to cut CI/CD times, facilitating rapid feature development. Additionally, worked on the development of ETL pipelines using MSSQL SSIS to efficiently collect and process customer response data. ### Research Intern @ Centre for Vision, Speech and Signal Processing at University of Surrey Jan 2021 – Jan 2021 Designed and implemented a new Zero-shot Sketch Based Image Retrieval method, that uses Meta-learning and test-time training to outperform prior methods by 0.17 mAP on standard benchmarks. Work presented as part of CVPR 2022. ### Research Intern @ International Institute of Information Technology Hyderabad (IIITH) Jan 2019 – Jan 2019 Worked on various computer vision and deep learning fields including image translation, semantic segmentation, and object detection. Focused on the segmentation of tables in degraded medical documents; created an annotated dataset and developed a semantic segmentation model baseline. Our work was published at GREC 2021 ### Intern-Product Labs @ CIE IIIT Hyderabad Jan 2018 – Jan 2018 | Hyderabad Area, India Developed market-relevant products based on technologies from IIIT-Hyderabad's research labs. My role involved understanding and implementing research papers to build prototypes. I also created a deep learning-based form parsing tool for rapid processing of handwritten forms. ## Education ### Master of Science - MS New York University ### Bachelor of Technology - BTech in Computer Science Mahindra University ## Contact & Social - LinkedIn: https://linkedin.com/in/vaishnav-potlapalli - Website: http://www.vaishnavrao.com --- Source: https://flows.cv/vaishnavpotlapalli JSON Resume: https://flows.cv/vaishnavpotlapalli/resume.json Last updated: 2026-04-05