Khasim Bin Saleh is a Student Researcher at Applied Perception and an MSc Artificial Intelligence student at SRH Germany University of Applied Sciences. He builds LLM and machine learning systems for settings where mistakes matter, and evaluates them before trusting them.
His MSc thesis, in progress, is Agentic Privacy Auditing of Synthetic Health Data: Evaluating LLM Audit Agents against Attack-Based Re-identification Metrics. As a research assistant at SRH in 2026 he compared industry surveys on enterprise AI agent adoption and matched agent design patterns (ReAct, Reflection, Planning, Tool use, Multi-agent) to real deployments. He is building Klindok, an offline clinical documentation tool that keeps patient data inside the practice.
Earlier, as a machine learning intern at Coschool, he built an AI tutor that reached more than 1,000 beta users, with evaluation pipelines and retrieval-augmented generation. He holds a B.Tech in Computer Science and Engineering with a specialisation in IoT from Lovely Professional University, and is co-inventor on a patent application filed in 2023.