Harsh Kanani

I am a Computer Science (Data Science) undergraduate with hands-on experience building production-grade backend and AI systems, currently working as an AI/ML Intern at AVD Inovex. I enjoy working on problems at the intersection of distributed systems, developer tooling, and applied machine learning.
Through my projects, I focus on building end-to-end systems rather than isolated prototypes. In VESPER, I developed an evidence-grounded financial intelligence platform with retrieval transparency, observability, and cost governance. In Docsmyth, I worked on API contract discovery using AST analysis and GraphQL introspection, emphasizing correctness, validation, and developer experience.
Beyond systems work, I have also worked on data-centric AI projects. I published a Gujarati Sign Language dataset to contribute toward accessible and regional AI research. Working on dataset creation gave me a deeper appreciation for how much real-world AI performance depends on data quality, labeling rigor, and thoughtful problem framing.