Abdulmalik Adeyemo

I'm an AI Engineer and researcher, currently pursuing an M.Eng in Computer Science at Nagoya Institute of Technology. My work sits at the intersection of building and evaluating multi-agent LLM systems: I design and deploy them for real users, and I research the architectures and evaluation methods that make them reliable. WHAT I'M WORKING ON On the engineering side, I build production AI systems across healthcare, education, and decentralized finance. Recent work includes RSMS, a multi-agent mentoring platform currently in active use by research students at Ozlab (NITECH); an ambient AI agent system for student risk management with voice-AI outreach and real-time human-in-the-loop supervisor workflows; and Toly.AI, a natural-language analytics platform for Solana DeFi data. On the research side, my first-authored paper on contestable AI and supervisor-in-the-loop agent architectures has been accepted at the International Conference on Intelligent Tutoring Systems (ITS 2026), peer-reviewed and forthcoming in the proceedings. I'm currently extending that work into healthcare through CARE-Agent, an ongoing project on doctor-configurable post-consultation patient follow-up with tiered escalation, scope-tagged adaptive memory, and safety guardrails for clinical deployment. My research interests span multi-agent system evaluation, contestable and human-in-the-loop AI, LLM-as-Judge methodology, and adaptive memory architectures for trustworthy agents. EARLIER CAREER Before my current research focus, I worked as a data scientist across Nigeria, the UK, and South Africa, building medical imaging models, enterprise data warehouses, and high-volume SQL systems. I bring that engineering foundation into every research project I take on. BEYOND THE WORK I've mentored over 150 engineers and students, including a 30-day intensive bootcamp on building production LLM agents with the LangChain ecosystem, and spoken at GITEX Africa Morocco, the Digital Africa Conference, and TechGala Japan. I previously founded FundusAI, developing AI-powered tools for diabetes care in Africa. TECHNICAL STACK LangChain, LangGraph, LangSmith, Prompt Engineering, Context Engineering, Harness Engineering, RAG, Multi-Agent Orchestration, LLM-as-Judge Evaluation, Contestable AI, Next.js, FastAPI, PostgreSQL, Neo4j, Python, PyTorch, SQL, Docker, AWS, Git, CI/CD Open to research collaborations, technical discussions, and speaking invitations.