Hey, I am Ruchit

I am a Master's student in Computer Science at the University of Edinburgh, researching at the intersection of LLM inference and systems programming. I am building an AI coding verifier for compiled languages: a Docker-based sandboxed execution infrastructure for Go and Rust, with a compile-execute pipeline and structured RL-compatible diagnostic feedback, supervised by Dr. Luo Mai. Alongside my research, I consult for Loop AI, building multimodal wearable data pipelines and biological age prediction systems using XGBoost, QLoRA-fine-tuned LLMs, and SHAP-driven explainability.

Before Edinburgh, I spent 3.5 years at Kroger (via ZopSmart), rising from SDE 1 to Senior SDE/Technical Lead. I led the Merch Hub platform ($25M projected value), architected Identity Provider-agnostic communication across pillars, and built production-grade gRPC infrastructure from scratch — achieving 98% improvement in resource efficiency and 88% reduction in heap allocation. Earlier, I created a self-service Federated Identity Management platform that cut merchant onboarding time by 60%, and designed an in-house SSO solution on Azure AD and LDAP.

My technical interests sit at the intersection of LLM inference, distributed systems, and low-level systems programming. Go and Rust are my primary tools. I write about what I'm learning and building here. Outside of code, I spend my time bouldering (currently at V5) and covering songs from Lana Del Rey to Metallica on my Fender Stratocaster and Gibson EC Custom.