# Shalin Shah > AI/LLMs @ Glean | Ex-Meta | CMU CS/AI Location: Menlo Park, California, United States Profile: https://flows.cv/shalinshah ## Work Experience ### Software Engineer (Machine Learning) @ Glean Jan 2025 – Present | Palo Alto, CA AI org, Glean Assistant (Agents + Chat) ML Team. Applying LLMs to build "Work AI for all". Projects include: - Agentic File System (AFS): Built + shipped AFS system for Glean Chat: in-house cloud-based agent sandboxes with shell tool access to file system, programmatic tool calling, and skill management, to drastically accelerate co-working product offering. - Code Interpreter/Writer: Took full ownership to drive performance + quality on data/file analysis queries in Glean Chat: built + shipped new agentic reasoning/looping system (PyAgents) atop OpenAI Code Interpreter APIs + Glean Context/Indexing, reducing latency by 40% and user-facing errors by 70%. Also optimized UX (with PM + Design collaboration). - Query Classification (QC): Designed + built + shipped E2E LLM-based QC system to run on ALL production Glean Chat queries, yielding live user intent insights/trends and task-specific evalsets, while optimizing LLM cost-quality tradeoff and internal UX. ### Software Engineer (ML and Backend) @ Meta Jan 2022 – Jan 2025 | Menlo Park, California, United States Facebook Video Ads Quality Team (01/2024 – 01/2025): - Analyzed user behavior/interaction data on FB Reels to design new quality bids, i.e personalized ad scoring/ranking terms incorporating user behavior ML predictions, to optimize user-engagement/revenue tradeoff. - Designed and built end-to-end new backend logic (Python, C++), data pipelines (SQL, Python), and production ML models (PyTorch + internal tools) for new quality bids, and experimentally verified engagement/revenue impacts & reliability to launch. - Developed LLM-based (Llama3) pipelines to augment ad classification data for use in quality bids. Instagram Reels Recommendation Retrieval Team (08/2022 – 01/2024): - Overall goal: drive user sessions & watch time, by improving candidate retrieval stage of IG Reels recommender system/algorithm. - Identified, implemented, and integrated cutting-edge ML methods from recent papers (with PyTorch + internal tools). - Designed & implemented new approaches for rule-based candidate-reel scoring/filtering, in backend (Python/Django) codebase. ### Fellow @ 8VC Jan 2021 – Jan 2021 Engaged in an immersive three month program to meet and learn from entrepreneurs, executives, and investors at Silicon Valley startups. ### Product/Software Intern @ Ikigai Jan 2021 – Jan 2021 - Innovated new product feature (NLP-driven data pipeline generation), and applied product management principles (competitive positioning, adoption barriers, etc) to develop specs/plan, working directly with the CEO of Ikigai. - Towards this, built (Python) new beginner-friendly search function for data operations in our platform, with NLP methods (semantic similarity, lemmatization, etc). ### Researcher @ Carnegie Mellon University School of Computer Science Jan 2021 – Jan 2021 Researched techniques to improve 3D Object Detection for autonomous vehicles by fusing LIDAR and RADAR data, in Deep Convolutional Neural Network (CNN) methods, in the Kitani Lab at CMU. ### Software Engineering/Data Science/Bioinformatics Intern @ GRAIL, Inc. Jan 2020 – Jan 2020 | Menlo Park, California, United States Designed and implemented evaluation framework and additional functionality for a proprietary statistical machine-learning model for genomic/epigenomic data, at a key step in the company's bioinformatic pipeline, in Go. ### Research Intern @ Stanford University School of Medicine Jan 2019 – Jan 2019 | Stanford, CA Research Internship in Rhiju Das Lab, Department of Biochemistry, Stanford University (School of Medicine), Summer 2019. - Developed comprehensive, user-friendly, flexible, and well-documented Python package for analysis of M2-Seq data, enabling researchers to run the entire bioinformatic pipeline, from raw sequencing output to annotated RNA structure diagrams, in just one command. - Implemented (in Python) and optimized M2-Net algorithm (essentially a simple Convolutional Neural Network) to pick out RNA helices from Z-scores of reactivity matrices, as part of the package. - Tools used: Python, Python Packaging, numpy, argparse, logger ### Software Engineering and Bioinformatics Intern @ Clear Labs Jan 2019 – Jan 2019 | Menlo Park, CA - Took primary responsibility for ML project to use Recurrent Neural Networks (RNNs) with LSTM to improve categorization of raw DNA sequence data. Model deployed in production. Tools: Python, TensorFlow, Keras, LIME, pandas, numpy, matplotlib. - Implemented thorough testing for a core backend service (data pipeline definition). Tools: Java, Spring, SQL, gRPC + microservices. ### Research Intern @ Stanford University School of Medicine Jan 2017 – Jan 2018 | Stanford, CA Research in Computational Genomics, Greenleaf Lab, Department of Genetics, Stanford University (School of Medicine). Full-time - Summer 2017, Part-time - Fall 2017 and Spring/Summer 2018. Investigated contribution of DNA-topology variations to polygenic disorders (including Rheumatoid Arthritis and Multiple Sclerosis), by applying new computational techniques (including LD Score Regression Algorithm) to large genomic/epigenomic datasets. Discovered significant link between disruption of cohesin (SMC3) binding sites on DNA and Rheumatoid Arthritis. ### Founder and President @ RoboKnights Institute of Inspired Learning Jan 2015 – Jan 2018 RoboKnights Institute of Inspired Learning is a 501(c)3 nonprofit organization that works to inspire curiosity and creativity in grade-school students through robotics teaching initiatives. The most notable of these are the FIRST Robotics Gujarat Initiative and the MSJE FLL Robotics Initiative, which have trained over 1000 students since 2015. Please see “Volunteer Experience” section for more details. ### Team Captain and Chief Software Engineer @ RoboKnights - FIRST Tech Challenge Robotics Team #5220 Jan 2011 – Jan 2018 | Fremont, CA - Attracted talented engineers from 3 World-Championship participant teams - Led engineering-brainstorming, goal-setting, work-planning, and time-commitments across 11-member team -Single-handedly developed award-winning robot software (autonomous and remote-controlled) for 5 years, in Java (with Android Studio) and C. Example codebase (from 2015-2016 season): https://github.com/RoboKnights/RoboKnights_Repository -Trained team, developed software peer-recognized as most advanced at World Championship, included dynamic position tracking, optimal route planning, computer vision. - Advanced to World Championship, 2014, 2016, 2017, 2018 - Top 10 in World Robot Performance, Top 6 in World Outreach and Public relations, 2015-2016 - Semifinalist Alliance (1 of 4), World Championship, 2017 and 2018 ### Chess Teacher @ Mission San Jose Elementary School Chess Team Jan 2013 – Jan 2016 | Fremont, CA Taught national-championship-winning elementary school chess team. Responsible for training a 20-student subgroup under Head Coach Joe Lonsdale Sr. ## Education ### Bachelor of Science - BS in Artificial Intelligence (School of Computer Science) Carnegie Mellon University ### Stanford Online High School ### Homeschool ## Contact & Social - LinkedIn: https://linkedin.com/in/shalin-s --- Source: https://flows.cv/shalinshah JSON Resume: https://flows.cv/shalinshah/resume.json Last updated: 2026-04-11