I'm Maahir Garg, and I build careful systems for unruly data.

Built an internal Claude Code-style CLI agent from scratch on Chainlit with modular tool-calling and a skill-routing system over proprietary models. Adopted by 100% of my team as the primary AI interface where classified-data constraints rule out Copilot. Also deploying agentic LLM pipelines that automate internal audit workflows end-to-end; two previously manual data sources now run effectively instantly.

Co-inventor of a patent-pending multimodal tracking framework for hand–eye coordination assessment in stroke rehabilitation, unifying Apple Vision Pro and iPhone into one tracking surface at 89% cross-device accuracy (Swift, RealityKit). Ran controlled user studies with healthcare practitioners and iterated system design from empirical findings. Paper in preparation.

Three semesters across CS3230 (Design & Analysis of Algorithms), CS2040S (Data Structures & Algorithms), and CS1231S (Discrete Structures). Coached students through dynamic programming, graph algorithms, and complexity analysis. 4.8/5.0 teaching effectiveness rating vs 4.2 faculty average.

Mentored 12 student engineering teams on software architecture, implementation planning, and release management. All 12 projects passed at or above their targeted grade level.