iHUB DivyaSampark, IIT Roorkee Launches Post Graduate Certificate in Agentic and RAG Systems for AI Engineering
The next wave of enterprise AI will not be built by professionals who can use AI tools. It will be built by engineers who can design, deploy, evaluate, and maintain AI systems that hold up under real conditions. As Retrieval-Augmented Generation and Agentic AI move from research into production, organisations face a growing shortage of engineers who can take AI beyond the demo.
Recent research highlights the scale of this gap:
• 92% of companies plan to increase their AI investments over the next three years, yet only 1% of leaders describe their deployments as mature. (Source: McKinsey, Superagency in the Workplace, January 2025)
• 33% of all enterprise software applications will include Agentic AI by 2028, up from less than 1% in 2024. (Source: Gartner, June 2025)
• 60% of companies name the scarcity of AI engineering talent as their single biggest barrier to transformation. (Source: McKinsey, Tech Talent Gap, March 2025)
Recognising this critical gap, iHUB DivyaSampark, IIT Roorkee has launched the Post Graduate Certificate in Agentic and RAG Systems for AI Engineering, a six-month, 100% live online programme designed to equip working technology professionals with the hands-on engineering skills needed to build, deploy, and manage production-ready AI systems.
Unlike programmes that focus primarily on AI concepts or prompt engineering, this programme is built around the full AI engineering lifecycle. Intermediate familiarity with programming concepts is expected through prior coding experience. Participants learn to:
• Build advanced RAG pipelines with hybrid retrieval, reranking, and evaluation using RAGAS and DeepEval
• Develop autonomous Agentic AI systems using LangGraph and ReAct architecture, with failure handling and human-in-the-loop workflows • Fine-tune open-source models using LoRA and QLoRA
• Deploy AI systems using FastAPI and Docker
• Monitor production AI systems using Langfuse and OpenTelemetry
• Design for AI governance including security, guardrails, MCP integration, PII detection, and India’s DPDP compliance
“Enterprises are deploying AI faster than they can find engineers who know how to build it reliably. This programme is our response to that gap. We have designed it around the specific engineering skills that production AI systems demand, not the concepts that look good in a syllabus. Every module, every project, and every tool reflects what engineers actually need to build AI systems that hold up in the real world.” — Prof. Gaurav Kumar Nayak, Assistant Professor, IIT Roorkee
The curriculum is structured across six integrated learning pillars: Python for AI Engineering, LLM Engineering and Evaluation, Enterprise RAG and Retrieval Infrastructure, Model Adaptation and Fine-Tuning, Agentic AI and Autonomous Workflows, and AI Infrastructure, Security, Governance and Deployment. Participants work hands-on with 15+ industry-standard tools, including:
• LangGraph, CrewAI, and Smolagents for Agentic AI
• Qdrant, FAISS, and Chroma for retrieval
• Hugging Face, Unsloth, and Axolotl for fine-tuning
• FastAPI, GitHub Actions, and Langfuse for deployment and observability • Guardrails AI, NeMo Guardrails, and Model Context Protocol for governance
“The talent gap in production AI engineering is real and widening. Organisations are investing heavily in AI but struggling to find professionals who can take a system from prototype to production. This programme from iHUB DivyaSampark, IIT Roorkee is built to produce exactly those professionals, engineers who can own the full lifecycle of an AI system and deliver something their organisation can trust.” — Manish Anand, CEO, Emeritus India
A key differentiator of the programme is its emphasis on building rather than watching. Participants complete six module-end projects across every pillar of the curriculum, plus a final capstone where they deliver:
• A production-grade RAG pipeline with hybrid search and reranking • A LangGraph-orchestrated autonomous agent with tool-calling and failure handling
• A FastAPI deployment for scalable API serving
• A Langfuse monitoring dashboard tracking latency, cost, and answer quality in real time
• A GitHub repository with the complete system and a deployed API endpoint
An optional two-day on-campus immersion experience is also available at IIT Roorkee’s Noida campus.
Programme at a Glance
Programme: Post Graduate Certificate in Agentic and RAG Systems for AI Engineering Institute: iHUB DivyaSampark, IIT Roorkee
Duration: 6 months
Format: 100% live online, no career break required
Eligibility: Bachelor’s degree or higher. Diploma holders with 3 years of work experience are eligible. Intermediate programming familiarity through prior coding experience required.
Weekly Effort: Approximately 8 to 10 hours
Programme Fee: INR 1,20,000 + GST
Certification: Post Graduate Certificate from iHUB DivyaSampark, IIT Roorkee upon successful completion
