Build
Make LLM applications.
Use prompting, RAG and agents to build applications on your own data.
University Programmes · University · 4–6 semesters
Generative AI & MLOps. Build LLM applications, and run them well.
Rooman's Generative AI & MLOps track is part of the Credit-Based Integrated AI Course, which universities embed in B.E., B.Tech., B.Sc. and BCA degrees with formal credits. Over 4–6 semesters it covers large language models, RAG, agents, model serving, evaluation and observability, taught by Rooman trainers alongside your college faculty.
180 hours7 modulesCredit-basedCapstone project
What you'll be able to do
Take an LLM application to production.
Building a generative AI demo is quick; running one reliably is the hard part. This track teaches both, inside your degree and with credits as set by your university.
Serve
Put models into production.
Package, deploy and scale models behind an API.
Measure
Know when it works.
Evaluate outputs, monitor quality and cost, and trace failures.
Skills covered
- Python for AI applications
- How large language models work
- Prompt design
- Retrieval-augmented generation (RAG)
- AI agents and tool use
- Model serving and APIs
- LLM evaluation
- Observability and monitoring
Who it's for
Built for your first AI engineering role.
The track is for undergraduates at colleges that partner with Rooman who want to build with LLMs. You enrol through your college, not directly with Rooman: the course runs where your university has a partnership with Rooman.
Roles where these skills are used
- AI Engineer (entry level)
- Generative AI Developer
- MLOps Engineer (graduate)
- LLM Application Developer
- Machine Learning Engineer (associate)
Entry requirements.
- Enrolled in a B.E., B.Tech., B.Sc. or BCA programme at a partner college
- Chosen as an elective or integrated track, as your university's scheme allows
- Basic Python; no AI experience needed
Not sure which course fits? Take the 10-minute career assessment, or talk to a counsellor.
180 hours · 7 modules
A curriculum paced by semester.
Seven modules, from ML foundations to observability, with a project semester at the end.
01Python and ML foundations24 hrs
The programming and machine learning basics that generative AI builds on.
Topics covered
- Python for AI work
- Supervised learning basics
- Evaluation metrics
You produce: a trained and evaluated baseline model.
02Large language models24 hrs
How LLMs work and how to prompt them well.
Topics covered
- Transformers and tokens
- Prompt design and structured output
- Fine-tuning concepts
- Limits and risks of LLMs
You produce: a prompt-driven LLM application.
03Retrieval-augmented generation26 hrs
Grounding model answers in your own documents.
Topics covered
- Embeddings and vector search
- Chunking and indexing documents
- Building and tuning a RAG pipeline
You produce: a RAG application over a document set.
04AI agents26 hrs
Models that plan, call tools and take actions.
Topics covered
- Tool calling
- Planning and memory
- Multi-step agent workflows
- Guardrails and safety
You produce: a working AI agent with tools.
05Model serving26 hrs
Deploying models behind reliable, scalable APIs.
Topics covered
- Containers for models
- APIs with FastAPI
- Scaling, latency and cost
You produce: a model served through an API.
06Evaluation and observability24 hrs
Measuring quality and watching systems in production.
Topics covered
- Offline and human evaluation
- Tracing and logging
- Monitoring drift and quality
- Experiment tracking
You produce: an evaluation suite and monitoring dashboard.
07Capstone project semester30 hrs
A team project that ships and monitors an LLM application.
Topics covered
- Use case with a faculty guide
- Build and evaluate
- Deploy and monitor
- Final report and viva
You produce: a deployed generative AI application and report.
Tools covered
The tools you'll work with.
You build with the open-source tools used for LLM applications and MLOps.
| Tool group | Tools | What you use them for |
|---|---|---|
| Language | Python, Jupyter | Write every application |
| LLM frameworks | Hugging Face, LangChain or LlamaIndex | Build RAG and agent applications |
| Retrieval | A vector database such as Chroma or FAISS | Store and search embeddings |
| Serving and MLOps | FastAPI, Docker, MLflow | Deploy models and track experiments |
Capstone
Ship an LLM application. Then measure it.
In the final semester your team builds, deploys, evaluates and monitors a generative AI application, guided by Rooman trainers and your faculty.
Every module ends with a hands-on lab in your college.
What you finish with.
- A RAG application
- An AI agent with tools
- A model served through an API
- An evaluation suite and dashboard
- A deployed capstone application
Your credential
Credits toward your degree.
Credits toward your degree. The track counts toward your degree, with credits as set by your university.
Assessment follows your university's scheme, with Rooman trainers and your faculty assessing the work together.
What's included
Everything you need to finish.
- Taught by Rooman trainers alongside your faculty
- Scheduled to fit your academic calendar
- Labs every semester and a capstone project
- Credits counted toward your degree, as your university sets them
- A curriculum aligned with your university's academic council
- Placement and hackathon support through your college
Before you enrol
Questions future AI engineers ask.
Can't find your answer? Call 080 6945 1000 or WhatsApp us.
Does it count toward my degree?
Yes. It is a credit-based track, with credits as set by your university.
Is my college a partner?
Ask your department or placement office, or a Rooman counsellor. You enrol through your college, not directly with Rooman: the course runs where your university has a partnership with Rooman.
Can my university partner with Rooman?
Yes. Universities embed the course in their degrees, and a custom track can be co-created with the academic council.
Who teaches it?
Rooman trainers, alongside your college faculty.
How long does it take?
4–6 semesters, scheduled to fit your academic calendar.
How is it different from the AI & Machine Learning track?
AI & Machine Learning builds models from the foundations. This track focuses on building applications with LLMs and running them in production.
Do I need AI experience?
No. Basic Python is enough; the first module covers the machine learning you need.
How do I find out fees and batch dates?
Talk to a Rooman counsellor on 080 6945 1000 or WhatsApp +91 97390 86029. They will share current fees, payment options and upcoming batch dates.
Related courses.
See all 5 tracks
180 hrsCredit-basedAI & Machine LearningMachine learning, inside your degree.
180 hrsCredit-basedCyber SecurityDefend networks, with AI on your side.
180 hrsCredit-basedData AnalyticsFrom raw data to decisions, for credit.For universitiesBringing this to your university?Embed this track in your degree programmes with formal credits, taught with your faculty.Talk to usNext step
Start with a conversation.
Tell us your college, course and year, or your university, and a counsellor will explain how the tracks work.
- Call: 080 6945 1000
- WhatsApp: +91 97390 86029
- Email: hello@rooman.com
Course details reviewed on 25 September 2026.