About Bibin Prathap

My focus

With 12+ years across enterprise technology, data science, and AI delivery, I focus on the gap between an AI proof of concept and a system an organisation can responsibly operate: measurable evaluation, traceable evidence, clear security boundaries, and accountable human decisions.

What I do

I help healthcare organisations, enterprises, and government teams move beyond chatbot demos into grounded, auditable AI: knowledge graphs that preserve domain context, MCP servers that expose enterprise tools with controlled access, and agentic automation that can run within a client's environment. These architectures can support data-residency and regulatory requirements; compliance depends on each organisation's implementation and controls.

Selected work

• VeritasGraph — an open-source GraphRAG framework focused on multi-hop reasoning, explainability, and source attribution. Explore the open-source work.
• Public-service AI — government-scale delivery work in Abu Dhabi, described in sanitized form across the case studies.
• Production AI — implementation notes on evaluation, identity, observability, and governance in the article library.

Recognition & Extraordinary Ability (EB-1A / Form I-140 Approved)

• Original Scientific Contributions — Published peer-reviewed research in Medical AI and Enterprise RAG architectures.
• Microsoft MVP — recognized in the AI Platform category, with a focus on Azure AI Services. See the Microsoft MVP profile.
• Judging & Peer Review — Regularly evaluating the work of peers in prestigious AI and healthcare technology panels.
• UAE Golden Visa holder for specialized technology talent, a testament to high-level recognition.
• International Speaker — Presented “Architecting Enterprise AI” at the MCT Middle East Summit 2026 (Riyadh, in collaboration with Microsoft).
• Speaker, mentor and author on medical AI and enterprise architecture.

Speaking & Mentorship

I mentor founders and speak on building AI that regulated customers can actually trust — ethical, auditable and sovereign by architecture, not by policy poster.

• Mentor — 2026 IEEE Tech Summit: Ethical AI (Global Founders Showcase, Budapest) — coached AI founders on evals, source attribution, audit trails, data residency and human-in-the-loop. Read the takeaways: 6 Engineering Decisions That Make AI Actually Ethical.
• Speaker — MCT Middle East Summit 2026 — “Architecting Enterprise AI,” Riyadh, in collaboration with Microsoft.
• Selected Speaker - Abu Dhabi Machine Learning, Season 7 Episode 1 (September 21, 2026; event held at MBZUAI, Abu Dhabi). Selected for "Trustworthy Deep-Research Agents: Grounding Evidence with GraphRAG, Multi-Hop Reasoning & Source Attribution." I was unable to attend in person. A related YouTube presentation explores auditable research agents and VeritasGraph; it is not presented as a recording of an in-person session at the event. Watch the related presentation | Organizer's programme | Organizer's LinkedIn post.

Where I've worked

• Government Entity, Abu Dhabi — AI Strategy Lead, Data Science & BI Analytics (2021 – present).
• Tech Mahindra — Senior AI & Data Science Consultant (2019 – 2021).
• Earlier roles in enterprise software, ERP and mobile across the UAE and India.

How I work

• Outcome-first — every engagement starts with the business metric we're moving.
• On-prem-friendly — I default to architectures that work without a public-cloud LLM API.
• Production discipline — evals, guardrails, observability and audit from day one.
• Knowledge transfer — your team owns the platform when I leave.

Get in touch

Read the enterprise AI solutions, browse case studies, or reach out directly.

Follow my work on LinkedIn, GitHub, and YouTube. Browse public coverage and speaking details on the media and recognition page.