Shreehitha Arushan

I’m a B.Tech graduate from IIT Kanpur (Materials Science & Engineering) with 2+ years of professional experience across software development and AI engineering. I focus on shipping end-to-end systems while keeping evaluation, reliability, and product quality tightly connected.

At Handshake AI, I work on LLM evaluation by authoring and hardening benchmark tasks across multiple review dimensions. I’ve reviewed and debugged benchmark task pull requests, including de-identification pipelines, and I’ve helped catch issues such as verifier exploits, contradictory instructions, and incomplete test coverage.

Previously at Outlier AI, I designed high-difficulty programming problems for the Spider Web project by analyzing real-world model failure patterns. As a senior reviewer, I authored gold-standard solutions, rubrics, and evaluation criteria, and validated peer submissions to maintain benchmark quality across the pipeline.

I also enjoy building full-stack products—like a personal finance portal with dynamic analytics, a robo-advisor with live market data and sentiment-driven strategies, and real-time stock tracking platforms. Whether it’s REST APIs, interactive dashboards, or automated code evaluation platforms, I bring a builder’s mindset and a problem-solver’s discipline to every project.