Jan 2022 – Aug 2024 · 2 yrs 7 mos

• Lead Generative AI R&D and Infrastructure
• Led the development of proprietary Generative AI models, including a Text-to-Speech (TTS) engine for Indic and global languages, and high-fidelity Lipsync technology.
• Architected a scalable AI infrastructure on AWS (Lambda, ECS, SageMaker, Batch) to power our custom in-house models, enabling the platform to generate over 2 million personalized videos daily.
• Team Leadership and Business Growth
• Recruited, scaled, and led a stellar team of 25 engineers, attracting top talent from leading global universities (IITs, BITS, CMU, Stanford) and tech institutions (Microsoft, Meta) to build and execute the product.
• Grew the company from $0 to over $1M+ ARR by translating the vision into a compelling product and user-experience.

Jul 2021 – Jan 2023 · 1 yr 6 mos · Mountain View, California, United States

• Serverless Compute Platform :
• Contributed to the development of the Databricks Serverless SQL platform deployed across AWS, Azure, and GCP.
• https://www.databricks.com/blog/2021/08/30/announcing-databricks-serverless-sql.html
• Enhanced container orchestration and allocation mechanisms, significantly reducing warm-up times and improving the efficiency of spark cluster allocation.
• Designed and implemented Fair Scheduling and Rate Throttling policies to streamline warm-pool cluster allocation, ensuring resource optimization and fair usage.
• Worked extensively with Kubernetes (AKS, EKS, GKE), Docker, Bazel, Prometheus, Scala, and more to deliver high-performance, scalable solutions.

Jan 2021 – Present · 5 yrs 4 mos

• Help building core AI infrastructure and creating AI-native products that win over customers.
• Guidance in building high-performing engineering teams.

Sep 2018 – Dec 2018 · 3 mos · Palo Alto, California, United States · Hybrid

• Provided remote academic support for Stanford’s core Machine Learning (CS229) and Deep Learning (CS230) courses.
• Held online office hours and advised students on course concepts, assignments, and project work.
• Assisted with grading and detailed feedback to reinforce student learning.

Aug 2018 – Jul 2021 · 2 yrs 11 mos · Palo Alto, California, United States

• NAS/Fileset Team:
• Accelerated Large-File Ingestion: Solved the long-tail problem for multi-TB-scale files by sharding data across multiple nodes, enabling high-speed ingestion.
• Ensured Backup Integrity: Led the design and implementation of a snapshot-verification service to validate backup integrity and provide auto-remediation for specific inconsistency scenarios.
• Optimized Storage Performance: Collaborated with a cross-functional team to enhance the data storage layer format, achieving a 3–4x performance improvement for large workloads.
• Boosted Metadata Scanning: Improved metadata scan performance for snapdiff-based NAS backups by 2–3x, delivering ultra-fast backups for large-scale workloads.
• Continuous Data Protection (CDP) Team:
• Achieved Near-Zero RPO: Enabled near-zero Recovery Point Objective (RPO) protection for VMware workloads by building high-performance data ingestion pipelines.
• Developed High-Speed Data Receiver: Designed and implemented the data-receiver service to ingest high-speed streaming I/O data efficiently.
• Built Multi-Tiered Buffering: Designed a multi-layered tiering buffer (Memory → SSD → HDD) to handle brief bursts in I/O traffic, ensuring seamless data capture.
• Real-Time Data Replication: Implemented a real-time replicator service to stream data to remote clusters with minimal latency.
• Enabled Resync Functionality: Designed and implemented resync feature to recover quickly from brief network outages, slowdowns, or I/O bursts, ensuring CDP system remains in sync.

Jul 2017 – Sep 2017 · 2 mos · Palo Alto, California, United States

• Distributed Systems | Scala | Cloud Computing
• Worked in a distributed system environment using Java and Scala to design a framework to allow Rubrik jobs to request compute instances in the cloud (Amazon EC2)

Jan 2017 – Jun 2018 · 1 yr 5 mos · Palo Alto, California, United States · On-site

• Served as a Teaching Assistant for CS110 (Principles of Computer Systems), CS229 (Machine Learning), and CS246 (Mining Massive Data Sets) during my Master’s at Stanford.
• Conducted office hours, discussion sessions, and occasional lectures for undergraduate and graduate students.
• Assisted in grading assignments and projects, providing detailed feedback to support learning.
• Mentored students on course concepts and project implementation, helping them tackle complex technical challenges.

May 2015 – Jul 2015 · 2 mos · South Korea

• ROS & SLAM Expertise: Developed a strong understanding of Robot Operating System (ROS) and explored Simultaneous Localization and Mapping (SLAM) techniques for map building.
• Autonomous Exploration: Enabled autonomous exploration capabilities in AGVs for efficient and accurate map building.
• Multi-Sensor Integration: Designed an interface to integrate open-source laser-based SLAM with Kinect camera images, leveraging 3D depth information for enhanced mapping accuracy.

May 2014 – Jul 2014 · 2 mos · Boston, United States of America

• Machine Learning & Hardware Design: Designed and implemented a custom hardware architecture for efficient feature extraction from images.
• Human Body Part Classification: Implemented decision forests to classify human body parts from depth images captured by a Kinect camera.