Jigyasa Gupta

Software Engineer – Practicum Project at Snap Inc.

Built a native C++ desktop streaming app for Snap AR Spectacles, enabling wireless connection to the host computer and streaming 30+ FPS in an immersive virtual environment. Implemented Bluetooth input with latency under 100ms for remote control. Designed and integrated OpenXR graphics framework for a responsive virtual desktop at 1200x800 pixels. Authored SDLC-related technical documentation and collaborated with 4 Snap engineers to address developer challenges and risks.

Software Dev Engineer Intern at Amazon Web Services

Deployed MCP server support on AWS Labs for RDS PostgreSQL and Aurora PostgreSQL with 2 open-source LLM tools; established a connection pool and drivers for Postgres for multi-client usage with 30 concurrent connections. Engineered an automated pipeline for AI-assisted coding, testing and code reviews for new open-source commits, monitoring uptime and optimizing prompts for code quality, maintaining a CI/CD pipeline and checkpoints for query plan management tests to improve developer productivity by 40%.

Associate System Architect Intern at Pegasystems Inc. (Pega)

Led a team to develop a Cross-Platform Key Management app; Developed a full-stack Canteen Management application; Simplified payment integrations across mobile and web; Achieved milestones of Certified System Architect and Senior System Architect within 2 months; Adopted Agile methodologies for iterative development.

Research Intern at Defence Research and Development Organisation (DRDO)

Initiated Digital Twin modeling of the human gut using explainable AI and BERT LLM; Evaluated estimator performance across 10+ datasets increasing gut-absorption model prediction accuracy to 94%; Led an intern initiative of 10 members to model interpolation and visualization of oxygen plots at high altitudes, improving physiological monitoring by 26%.

Open-Source Contributor at Google Summer of Code (National Research for Network Biology)

Enhanced support for Pathway Commons Database (v12) in clusterProfiler, increasing user base by 72%; Collaborated with 8 global contributors to implement data analysis methods, improving package reliability and scalability by 40%; Integrated advanced debugging scripts for data science; Addressed and refactored code to resolve 40+ open-source issues, enabling 48% faster visualizations.