AI Expert · 3 months · 90 days · 360 hours · ~6 hrs/day in person
AI FORWARD · BUILD AI · TIER 3 · BENGALURU RESIDENCY

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
Course Duration

360 Hrs

Delivery Mode

In-Person · Bengaluru residency

Certification

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

Minimum

10 LPA

Average

16 LPA

Maximum

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

LangGraph CrewAI MCP OpenAI Agents SDK Claude Agent SDK RAG MLflow Triton FastAPI Kubernetes Cursor Claude Code

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
Enrol Now →

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.

Enrol Now → Email an advisor