AI Expert
Pod-Ready In-Person Production
100% Placement Assistance · 1,000+ Hiring Partners · Rooman + NSDC Recognition
- Industry-Relevant Curriculum
- Hands-On Training
- Experienced Instructors
- Placement Assistance
- Live Real-Time Projects
360 Hrs
In-Person · Bengaluru residency
Rooman AI Engineer Certificate + capstone portfolio + product engineering lead recommendation
Course Overview
AI Engineer is Tier 3 of the Build AI pipeline — the bridge between AI Professional and a real client pod. A 360-hour in-person residency at Rooman's Bengaluru campus. Seven modules covering production agentic systems, RAG mastery, AI-augmented engineering, client-side delivery skills, and AI security & governance. The capstone is unique — 4 weeks embedded inside a real Rooman product engineering team (Omnis, BlueLinked, AI Tutor, or CRM Voice Agent). Graduates exit pod-ready, with three documented placement pathways: Rooman GCC pod placement, StartupVarsity Founders' Track, or hiring partner network. Selection-based — approximately 1 in 3 applicants accepted.
Key Features
Duration
3 months · 90 days · 360 hours · ~6 hrs/day in person
Mode
In-Person · Bengaluru residency
Projects
Real production capstone (4-week embedded engineering inside Omnis / BlueLinked / AI Tutor / CRM Voice Agent)
Certification
Rooman AI Engineer Certificate + capstone portfolio + product engineering lead recommendation
Placement Support
GCC pod placement / StartupVarsity Founders' Track / 1,000+ hiring partner network
Tools Covered
Production agentic stacks · MCP · RAG mastery toolchains · LangGraph · CrewAI · OpenAI Agents SDK · Claude Agent SDK · MLflow · Triton · FastAPI
Skills Covered
✓ Production agentic systems with failure recovery and human-in-the-loop
✓ RAG mastery — chunking, hybrid retrieval, reranking, evaluation
✓ AI-augmented engineering — senior-level Cursor / Claude Code workflows
✓ Client-side delivery — communication, async work, code review etiquette
✓ AI security and OWASP LLM Top 10 defence
✓ Production-grade AI deployment and observability
✓ Capstone embedded in real Rooman product team
✓ Pod-ready certification by product engineering lead
Course Curriculum
7 modules built end-to-end from fundamentals to capstone.
Module 1 — Production AI Engineering at Scale (60h)
- Cost engineering — token budgets, prompt caching, semantic caching
- Multi-model orchestration — routing, fallback chains
- Latency engineering — TTFT, streaming, p50/p95/p99 SLOs
- Gradual rollout — feature flags, canary deployments, A/B testing
- Regression testing — golden datasets, eval-on-commit in CI
- Production logging — what to log, PII handling, observability
- Capstone — harden a prototype for 10K-user/day at $/month ceiling
Module 2 — Agentic Systems Deep Build (70h)
- Agent taxonomy — reactive, deliberative, hybrid
- ReAct, Plan-and-Solve, ReWOO, Tree-of-Thought patterns
- LangGraph for state-machine agent design
- CrewAI for multi-agent orchestration
- OpenAI Agents SDK + Claude Agent SDK at depth
- MCP (Model Context Protocol) — server authoring + governance
- Failure recovery + human-in-the-loop patterns
- Hierarchical and long-horizon agents with memory
- Capstone — customer-support agent, 5 task types, $0.10/ticket
Module 3 — RAG and Retrieval Mastery (50h)
- Why most RAG implementations fail — diagnostic mindset
- Chunking strategies — fixed, semantic, hierarchical, late chunking
- Embedding model selection and benchmarking
- Vector databases — pgvector, Pinecone, Qdrant, Weaviate
- Hybrid retrieval (BM25 + dense), reranking with cross-encoders
- Query rewriting, HyDE, multi-query expansion
- Retrieval evaluation — precision@k, MRR, NDCG, Ragas
- Capstone — 100K-document RAG at 0.85+ precision@5
Module 4 — AI-Augmented Engineering at Senior Pace (40h)
- Spec-driven development — writing AI-executable specs
- Cursor and Claude Code at senior depth
- Code review when an AI is the contributor
- Test-first AI engineering — TDD with AI tools
- Multi-agent code workflows — planner + coder + reviewer
- When NOT to use AI tools — security paths, novel design
- Lab — ship a 4-day feature in 1.5 days, defect rate ≤ hand-written
Module 5 — Client-Side Delivery in International Pods (40h)
- International pod context — US/EU/UAE/SG client expectations
- Cultural calibration — Indian engineering norms vs client norms
- Written communication — PRs, design docs, RFCs at senior level
- Async working patterns — when to write vs call
- Working in client repos — fast codebase orientation
- Sprint planning across timezones without daily standups
- Live demos and client presentations
- Business English at engineering depth
- Lab — shadow a real Rooman GCC pod for one week
Module 6 — AI Security, Governance & Regulated Industries (30h)
- OWASP LLM Top 10 — prompt injection, output handling, supply chain
- Prompt injection defence — instruction hierarchies, output filters
- PII handling in LLM workflows — detection, redaction, audit logs
- GDPR for AI — lawful basis, right to explanation, automated decisions
- HIPAA for healthtech AI — PHI handling, BAA requirements
- SOC 2 Type II for AI systems — control mapping, audit-ready logging
- EU AI Act — risk categorisation, transparency obligations
- Adversarial robustness, red-teaming, abuse monitoring
- Capstone — red-team a Rooman product AI feature
Module 7 — Capstone Embedded in a Real Rooman Product Pod (70h)
- Pod assignment — Omnis / BlueLinked / AI Tutor / CRM Voice Agent
- Codebase orientation, dev environment, AI toolchain setup
- Sprint 1 — paired feature development with senior engineer
- Sprint 2 — independent feature delivery with code review
- Final demo — present shipped feature to product engineering lead
- Final assessment — pod-ready or needs-another-sprint verdict
- Placement match — GCC pod / Founders' Track / hiring partner
Salary Scale
10 LPA
16 LPA
30 LPA
Pod-ready AI Expert graduates command 30-50% premium over generic AI engineers. Top graduates placed into Rooman GCC pods serving US, UK, EU, UAE, and Singapore clients.
Job Roles You Can Target
Senior AI Engineer (pod-ready)
Production ML Engineer
AI Solutions Architect
Rooman GCC Pod Engineer
StartupVarsity Founding Engineer
AI Product Engineer
Eligibility Criteria
- AI Professional (Tier 2) completion — required
- Selection interview + portfolio review (approximately 1 in 3 applicants accepted)
- Strong English communication for client-side delivery work
- Commitment to 90-day in-person Bengaluru residency
- Engineering degree strongly preferred
Tools & Technologies
Training Options
Classroom Training
₹85,000 ₹65,000 Including Taxes*
- Certified Industry Expert Trainers
- AI-Powered LMS with 1-Year Access
- In-Person Mentorship & Doubt Solving
- Fully Equipped Labs & Collaborative Learning
- Campus-Like Environment with Networking
Why Join this Program
Earn a Job
Complete job assistance tailored to your career goals. Expert placement guidance.
Industry-Expert Trainers
Learn from seasoned trainers. Real-world insights that go beyond textbooks.
Industry-Relevant Tools
Hands-on exercises and projects with integrated lab access. Latest tools.
Structured Curriculum
Industry-vetted, expert-shaped curriculum. Career-ready from day one.
Integrated Gen AI Modules
Cutting-edge Generative AI modules aligned with emerging tech trends.
Interview Prep & Placement
Sharpen interview skills with practical training. Complete placement support.
Program FAQ
Is AI Engineer only in-person?
Yes. The capstone requires being embedded inside a real Rooman product team, which only works in person. The other 6 modules are also delivered in person on Bengaluru campus.
What's the selection process?
Application + portfolio review + 45-min technical interview + 30-min behavioural interview. Approximately 1 in 3 applicants accepted.
Where do graduates get placed?
Three pathways — Rooman GCC pod placement (serving US/UK/EU/UAE/SG clients), StartupVarsity Founders' Track (join senior-led startup team), or hiring partner placement (1,000+ companies). Top graduates command ₹15-30 LPA at first offer.
What if I don't pass the final assessment?
We extend by one 4-week supervised sprint to close gaps. Most learners reach pod-ready by end of extension.
Can I take AI Engineer without AI Professional?
No. AI Professional is a hard prerequisite. You need the production engineering depth before going pod-ready.
What's the integrated programme?
AI Foundation + AI Professional + AI Engineer can be taken as an integrated 6-month programme for ₹1,00,000 total (₹30K stipend during AI Expert capstone).
Ready to start? Next cohort opens soon.
Limited seats per batch. 100% placement assistance.