LinkedIn · Software Engineer
Feb 2026 – Present · Mountain View, CA
- Build agentic features for LinkedIn Hiring Assistant, serving 1.5M+ users, using OpenAI and open-source LLMs.
- Led migration of two critical resume-processing pipelines from Apache Samza to Flink, handling 350K+ resumes per day and 350M resumes in total.
- Saved approximately $91K per year through the migration, reduced pipeline memory by 45% and 11% (approximately 8 TB total), and retired 60 TiB of legacy Kafka capacity.
- Build LLM evaluation pipelines and run SFT and RL fine-tuning on Qwen models serving millions of requests daily.
- Tech stack: OpenAI; Open-source LLMs; Qwen; Apache Samza; Apache Flink; Apache Kafka; LLM evaluation; SFT; RL.
C3 AI · Associate Solution Engineer
Jul 2024 – Jan 2026 · Redwood City, CA
- Shipped 8+ enterprise GenAI applications for government, manufacturing, and steel clients using RAG with PGVector and Gemini/OpenAI models, driving $4M+ in revenue and $10M+ in customer savings.
- Built ingestion pipelines for structured MSSQL data and unstructured PDF and Docx documents, using multimodal chunking and indexing for accurate retrieval.
- Developed a data model connecting 4+ data sources and a data pipeline for a major U.S. steel manufacturing facility, deploying it within 6 weeks.
- Built enterprise AI applications with the Solution Engineering Team on the C3 Platform using JavaScript, Python, JSON, and C3typ.
- Tech stack: Python; PGVector; Gemini; OpenAI; RAG; MSSQL; C3 Platform; JavaScript; JSON; C3typ; RhinoJS.
ClaimBrite · Software Engineer Intern
Feb 2024 – May 2024 · Remote (CO)
- Built and launched a responsive, customer-facing website using Next.js and Tailwind CSS, with optimized performance, modern design, and mobile-first responsiveness.
- Collaborated with designers, product managers, and stakeholders in an agile development environment to collect user feedback and iterate on website features.
- Implemented design improvements informed by the feedback and feature iteration process to make the website more intuitive and user-centric.
- Tech stack: Next.js; Tailwind CSS.
Keck Medicine of USC · Application Developer
Oct 2022 – May 2023, Aug 2023 – May 2024 · Los Angeles, USA
- Worked with the Salesforce Development Team on Lightning Web Components and Apex for Norris Cancer Hospital and the Transfer Center.
- Wrote JavaScript scripts to automate manual tasks within the organization.
- Resolved issues and wrote automated flows with the SharePoint and Power Automate team.
- Supported marketing for the internship program.
- Tech stack: Salesforce; Lightning Web Components; Apex; JavaScript; SharePoint; Power Automate.
C3 AI · Solution Engineering Intern
Jun 2023 – Aug 2023 · Redwood City, USA
- Collaborated with the Solution Engineering Team to design and develop AI applications on the C3 Platform using JavaScript, Python, JSON, and C3typ.
- Built a predictive maintenance platform for 1,500+ sensors at a manufacturing facility, partnering with one of the largest private companies in the U.S.
- Built an ML model management system for 800+ models using Python and React, allowing users to customize hyperparameters and test models in real time through a user interface.
- Tech stack: C3 Platform; JavaScript; Python; JSON; C3typ; React.
Harvard Medical School (McLean Hospital) · Research Developer
Jul 2021 – Jul 2022 · Boston, MA
- Owned full-stack development of a HIPAA-compliant mental health iOS application in Flutter, with a secure and scalable backend on AWS.
- Implemented backend development, routing, authentication, and video hosting using DynamoDB, Elastic Beanstalk, Route 53, Cognito, and S3.
- Served as the project’s sole cloud-security point of contact, coordinating with the cloud security team and leading security and hosting services.
- Developed the mental health application for Dr. David Rosmarin, managing its backend and iOS implementation.
- Tech stack: Flutter; AWS; DynamoDB; Elastic Beanstalk; Route 53; Cognito; S3.
METX Robotics · Research and Teaching Assistant
Apr 2019 – Jun 2021 · Mumbai, India
- Designed and developed Amyra, a service robot, and led its software architecture and implementation; the project was showcased at NESCO and Grand Hyatt in Mumbai.
- Programmed Arduino-based obstacle avoidance systems using ultrasonic and IR sensors, integrating software with embedded hardware.
- Worked on line-following robots and IoT-based automation systems, and authored a research thesis on a PID-based line-following algorithm.
- Taught over 20 students in STEM education and about 12 students in C programming.
- Managed a three-day event for METX Robotics at the NESCO Exhibition Centre.
- Tech stack: Arduino; C/C++; ultrasonic sensors; IR sensors; PID-based line following; IoT automation.