A 19-year-old developer from Kerala, India. I build things, break things, and occasionally fix other people's things too.

• B.Sc. Computer Science @ Sacred Heart College, Kerala (GPA: 3.81)
• Invitee to Hack Club Neighborhood 2025 in San Francisco
• Open Source Contributor to Google Gemini CLI and Oppia
• Interested in software engineering, AI developer tools, and full-stack development

Selected for Hack Club Neighborhood in San Francisco by building a procedurally generated browser game in React with on-demand room generation, randomised door layouts, and a unique exit per run, completing 130 hours of project work.

Contributed multiple improvements to Google's Gemini CLI, including:

Fixed a runtime crash affecting the VS Code extension. Improved GitHub repository setup by validating fork ownership, preventing cloud-environment failures. Added automatic skill activation through regex-based prompt matching with precompiled patterns for improved performance.

Notable PRs:

google-gemini/gemini-cli#22763 google-gemini/gemini-cli#18099 google-gemini/gemini-cli#17478

Contributed UI improvements to Oppia's Angular frontend, including:

Fixed contributor tooltip rendering so all contributor names remain visible and clickable. Added a responsive search-panel toggle for the blog homepage, improving usability while maintaining test coverage.

Notable PRs:

oppia/oppia#25001 oppia/oppia#25253

Built an AI-powered assistant that generates context-aware replies for X posts using Ollama by developing a Playwright-based tweet scraping pipeline that collects tweet data and performance from the feed, which uses a structured retrieval dataset of high-performing tweets and engagement metadata to improve response relevance and consistency.

Native Android expense tracker built with Kotlin.

Reads SMS transaction messages automatically. Extracts merchants and amounts using regex. Categorises spending into groups such as food, bills, shopping, and more.

Repository: https://github.com/019-Abhi/Gringotts-Expense-Calculator

Built during the MLH Midnight Hackathon.

A decentralized medical records platform where patients control encrypted health records stored on the Midnight blockchain, ensuring data remains private unless explicit access is granted.

Repository: https://github.com/Private-Health-Record

Languages Kotlin · JavaScript · TypeScript · Python · C++ · SQL · HTML · CSS Frameworks React · Flask Tools Git