Generative AI Course: Live Online Training with 6 Projects
Building your first application with a language model is one thing. Understanding why its answers fail, how to improve retrieval, or when fine-tuning makes sense is another.
At School of Core AI, our Generative AI course helps you develop the skills to work through those problems. Over 24 weeks, you'll learn how modern generative models work, build RAG applications, adapt language models with LoRA and QLoRA, explore multimodal AI, and evaluate the systems you create.
Designed for software developers, ML practitioners, and technical professionals who want to move beyond experimenting with AI models and develop practical Generative AI engineering skills.
Course at a Glance
24 Weeks
Live online classes
Live online
Instructor-led
6 Guided Projects
Defined deliverables
Instructor Feedback
Project reviews
SCAI Certificate
On completion
Fee: ₹64,999
Confirm applicable taxes and payment terms with admissions.
What Is This Generative AI Course and Who Is It For?
The School of Core AI Generative AI course is a 24-week, instructor-led online training programme focused on large language models, retrieval-augmented generation (RAG), model fine-tuning, multimodal applications, and LLM evaluation.
It is designed for people who want to understand how Generative AI systems are built and improved, rather than only learning how to use AI tools.
You might already know how to call an LLM API. Perhaps you've built a chatbot, experimented with Hugging Face models, or connected a vector database to a small application.
But building a dependable AI application introduces different questions.
Why does retrieval miss important information? When does fine-tuning make sense? How can you tell whether a model improvement is real? And how do you evaluate an answer that sounds convincing but isn't supported by the available information?
These are the kinds of problems this programme helps you work through.
Generative AI Training for Software Developers
If you already build applications, this course helps you understand the model and retrieval layers behind AI features.
You'll learn how to work with document collections, improve the information supplied to language models, adapt models for specific tasks, and evaluate generated outputs. The goal is to help you make better engineering decisions when adding Generative AI to an application or building a new one.
Generative AI for ML Engineers and Data Scientists
If you already work with machine learning, the programme takes you further into language models, parameter-efficient fine-tuning, multimodal systems, and GenAI-specific evaluation.
You'll compare model behaviour, create meaningful baselines, and identify whether changes genuinely improve a system.
Can You Join Without Generative AI Experience?
Yes, provided you have working knowledge of Python and basic technical concepts.
You don't need previous experience with RAG, LoRA, or language-model deployment. The early part of the programme introduces the foundations needed for the later projects. However, if you're completely new to programming, we recommend learning Python fundamentals first.
Which SCAI Course Should You Choose?
SCAI runs several related programmes: the AI Developer Course for building AI-powered applications, this Generative AI Course for deeper work in LLMs, RAG, fine-tuning, multimodal AI and evaluation, the Agentic AI Course for stateful agents, tools and orchestration, and the LLMOps Course for serving, infrastructure and operational reliability. The right programme depends on what you want to build and the skills you already have — not simply which course lists the most tools.
Compare them in detail in the course comparison below.
UNDERSTAND
Transformers & GenAI Models
GROUND
RAG & Retrieval
ADAPT
LoRA / QLoRA
EVALUATE
Quality & Failure Analysis
INTEGRATE
Serving & Applications
- Mode
- Live online cohort with recordings and project review
- Duration
- 24 weeks / 6 months
- Fee
- ₹64,999. Confirm applicable taxes and payment terms with admissions.
- Audience
- Developers, ML practitioners and technical working professionals
- Projects
- 6 guided systems
- Certificate
- School of Core AI course completion certificate
Learner Experiences and School of Core AI Reviews
These learner experiences are already published by SCAI under its GenAI Specialization. They describe individual learning experiences across different professional backgrounds.

Senior Data Scientist
I joined mainly to go deep into transformers and their variants. The specialization helped me understand LLM systems from the inside out — not just how to use them, but how to reason about and architect them.
GenAI Specialization
Praveen
Senior Software Developer
I wanted to learn AI development and agentic AI — how to build and deploy intelligent applications end-to-end. The track gave me hands-on exposure to the full pipeline, from idea to production.
AI DeveloperRead School of Core AI Reviews on Google
Read independent reviews of School of Core AI on Google. The business profile covers the institution and its programmes; it is separate from the learner experiences shown above.
Read reviews on Google(opens in a new tab)Who Should Join This Generative AI Course?
This course is best suited to software developers, ML practitioners and technical working professionals who know basic Python and want deeper capability in RAG, model adaptation, multimodal AI and evaluation. It is not positioned as a no-code AI-tools course for non-technical beginners.
Generative AI Course for Software Developers
Want to move beyond API integration into retrieval quality, model adaptation and evaluation.
Background: Basic Python, some backend or app experience
Gains: RAG, fine-tuning, multimodal AI and evaluation capability
Generative AI Course for ML Engineers and Data Scientists
Want deeper RAG, fine-tuning and evaluation skills with structured project evidence.
Background: Python, basic ML concepts, comfort with notebooks
Gains: Production-grade GenAI systems with evaluation evidence
Generative AI Course for Working Professionals
Want a live online format with recordings and project review while working full-time.
Background: Technical background with Python fundamentals
Gains: Portfolio-ready GenAI systems and career support
Is This Generative AI Course Suitable for Beginners?
If you have basic Python skills and are new to Generative AI, the refresher phase helps you get up to speed. If you do not yet know Python functions, modules and virtual environments, you should build those foundations first — this course does not teach programming from scratch.
Good fit if you have
Basic Python skills
Technical mindset
Ready for project work
Interest in evaluation depth
Generative AI Course Eligibility and Prerequisites
You should understand Python functions, modules, virtual environments and basic machine-learning terminology. No prior Generative AI experience is required, but you should be comfortable writing and running Python code.
This course is not ideal if you
Want only no-code AI
Want only prompt templates
Do not want programming
Expect guaranteed placement
What You Will Learn in Our Generative AI Training
A useful Generative AI application needs more than a model that produces fluent text. It needs the right information, appropriate model behaviour, reliable evaluation, and an application that can use the output.
The course connects these parts so you understand how individual techniques contribute to a complete system.
Understand Large Language Models and Transformers
You'll begin by exploring how generative models process inputs, represent context, and produce outputs.
This includes transformer architectures, tokenization, attention, neural-network fundamentals, and the behaviour of language and vision models. The purpose is practical: to understand how models behave well enough to recognize limitations, select suitable approaches, and troubleshoot unexpected results.
Key Areas
Transformers · tokenization · attention · model behaviour
Engineering Decision
How does the model behave, and where do its limits show?
Build RAG Applications Using Your Own Data
Many useful applications depend on information that isn't reliably available from a model's existing knowledge.
Retrieval-augmented generation addresses this by finding relevant information from documents or other knowledge sources and supplying it to the model when generating a response. You'll work with document ingestion, chunking, embeddings, vector search, hybrid retrieval, and reranking. You'll also test whether the system retrieves the right information, whether its answers are supported by that information, and what happens when the documents don't contain an answer.
Key Areas
Ingestion · chunking · embeddings · hybrid retrieval · reranking
Engineering Decision
Is the answer actually supported by the retrieved information?
Fine-Tune Language Models with LoRA and QLoRA
Sometimes better instructions or additional context are enough to improve an AI application. Sometimes a model needs to be adapted to a particular task.
You'll explore parameter-efficient fine-tuning using LoRA and QLoRA, including preparing suitable data, training adapters, and comparing adapted models against a baseline. You'll also learn when fine-tuning may not justify its additional cost and complexity.
Key Areas
LoRA · QLoRA · baselines · regression awareness
Engineering Decision
Does the adaptation justify its cost and complexity?
Work with Text, Images and Documents
Generative AI isn't limited to text conversations.
You'll explore multimodal models and workflows that combine language with visual information, including documents containing images, tables, or scanned content. The focus is on building practical workflows, evaluating their accuracy, and recognizing when a model cannot reliably interpret an input.
Key Areas
Vision-language · document workflows · abstention
Engineering Decision
Can the system handle inputs it cannot reliably read?
Evaluate GenAI Applications Before You Rely on Them
A response can sound convincing and still be incorrect.
You'll learn how to create evaluation datasets, check retrieval quality, identify unsupported answers, compare model versions, and investigate failure cases. These skills help you decide whether a change actually improves an application or introduces new problems.
Key Areas
Evaluation datasets · grounding checks · failure analysis
Engineering Decision
Did the change actually improve the application?
Connect Models to Applications and Explore Agentic Workflows
You'll learn how AI capabilities are exposed through APIs and integrated into applications.
The programme introduces model-serving fundamentals, application integration, controlled tool use, and the foundations of agentic workflows. Advanced multi-agent orchestration and large-scale infrastructure operations are explored further in SCAI's specialist programmes.
Key Areas
Serving fundamentals · application integration · bounded tool use
Engineering Decision
Is the model usable outside a local notebook demo?
How You Choose Between Prompting, RAG, Fine-Tuning and Agents
Use prompting when the model already knows the required information and needs clearer instructions. Use RAG when answers depend on private, changing or citable knowledge. Use fine-tuning when repeated task behaviour must change. Use a bounded agent workflow only when the system must select and execute tools across multiple steps.
| Problem signal | Start with | What must be measured | Course location |
|---|---|---|---|
| The model knows the facts but returns inconsistent structure or style | Prompting and structured outputs | task pass rate, schema adherence and consistency | Transformer and LLM workflow |
| The answer depends on private, updated or source-cited information | RAG | retrieval recall, citation correctness and grounded answer quality | RAG Knowledge System |
| The model repeatedly fails a specialised behaviour despite good context | LoRA/QLoRA fine-tuning | held-out task accuracy, format adherence and regression | Fine-Tuned LLM Assistant |
| The task depends on images, scans, charts or mixed documents | Vision-language or multimodal workflow | field accuracy, visual grounding and abstention | Multimodal Document Intelligence System |
| The task requires controlled API or tool execution across steps | Bounded agent workflow | task success, tool-call errors, step count and loop rate | GenAI Capstone |
| The system works but must be scaled, monitored and cost-controlled | LLMOps | latency, throughput, reliability, cost and production regressions | Dedicated LLMOps Course |
GenAI demos often appear successful because the output is fluent. Engineering review starts by testing the failure hidden behind that fluency.
| System | False sign of success | Failure learners detect | Review evidence |
|---|---|---|---|
| RAG | The answer sounds correct | required passage was not retrieved or the citation does not support the answer | retrieval metrics, citation check and query-level error analysis |
| Fine-tuned LLM | Training loss decreases | the target task improves but general or safety behaviour regresses | held-out set, baseline comparison and regression suite |
| Multimodal model | The response confidently describes an image | text was unreadable, fields were missed or visual evidence was invented | field accuracy, grounding check and abstention cases |
| Agent workflow | One happy-path demo completes | a failed tool call produces repetition, duplicated work or an uncontrolled loop | trace review, maximum-step control and task success rate |
| Served model | The local API returns a response | latency, memory or concurrency makes the system unusable outside the demo | basic latency/resource report; deeper operations continue in LLMOps |
SCAI GenAI Engineering Loop: define → select → build → evaluate → explain
Notebooks, datasets, lab prompts and solutions remain learner-only.
Six Generative AI Projects You'll Build
Learning a technique is useful. Applying it to a problem, testing the result, and explaining your decisions helps you develop real engineering confidence.
During the programme, you'll work through six guided builds. The first two strengthen model and evaluation foundations. The remaining projects apply those skills to retrieval, fine-tuning, multimodal AI, and an integrated GenAI system. Each project has defined deliverables and review criteria.
Model Training and Evaluation Workflow
What you build
Train and compare a baseline model
What you submit
Training script, evaluation report and baseline comparison
How it is evaluated
Reviewed for reproducibility, evaluation methodology and whether the comparison is valid
Engineering Question
How do you prove one model/version is actually better?
Transformer Behaviour Lab
What you build
Test tokenization, context and structured-output behaviour
What you submit
Lab notebook with documented tests and findings
How it is evaluated
Reviewed for test coverage, documentation quality and failure identification
Engineering Question
Where does the model fail despite good prompts?
RAG Knowledge System
What you build
Build ingestion, chunking, retrieval, reranking and citations
What you submit
Working RAG pipeline with retrieval metrics and citation checks
How it is evaluated
Reviewed for retrieval quality, citation accuracy and grounded answer correctness
Engineering Question
Does the system retrieve and cite the right evidence?
Fine-Tuned LLM Assistant
What you build
Adapt a suitable open-weight model with LoRA or QLoRA
What you submit
Fine-tuned model adapter, evaluation set and baseline comparison
How it is evaluated
Reviewed for task improvement, regression check and evaluation rigor
Engineering Question
Did adaptation improve the target without regressing safety?
Multimodal Document Intelligence System
What you build
Process text and visual document inputs
What you submit
Multimodal pipeline with field-accuracy results and failure cases
How it is evaluated
Reviewed for field accuracy, visual grounding and honest abstention on unreadable inputs
Engineering Question
Can the system handle unreadable or ambiguous inputs honestly?
Controlled GenAI Capstone
What you build
Combine retrieval, model access, tool use, API serving and tracing
What you submit
Integrated system with architecture decision record, evaluation set and failure analysis
How it is evaluated
Reviewed for integration quality, evaluation evidence and production readiness gaps
Engineering Question
Is the integrated system ready for production handoff?
What Makes a Strong Generative AI Project?
A working demonstration is a starting point.
A strong project also explains the problem, the design choices, the evaluation method, and what still needs to improve. For example, when building a RAG application, you should understand more than whether it can answer a handful of questions. You should be able to inspect retrieval failures, check whether source citations support an answer, and explain your retrieval strategy. That's the kind of thinking we want learners to practise throughout the programme.
How Generative AI Projects Are Evaluated
A working demo is not considered complete merely because it produces an answer. Each major project must include a defined task and baseline, an architecture decision record, an evaluation set, a failure analysis and a portfolio-safe system walkthrough.
BUILD
Define task, baseline, architecture
TEST
Run evaluation set and failure cases
EVALUATE
Quality · Grounding · Reliability · Safety
PASS / ITERATE
Document evidence or iterate
Try a Free GenAI Evaluation Exercise
Work through a short example using sample policy documents. Check evidence, identify an outdated answer, decide when to withhold an unsupported answer, and compare your reasoning with worked solutions. The reading exercise requires no paid API or GPU.
Meet the SCAI Mentors
Meet members of the SCAI mentor team whose published areas of work include language models, multimodal AI and machine learning. Teaching and project-review responsibilities depend on the cohort; confirm your instructor and review schedule before enrolling.
Niladri Bihari Das
Lead AI Engineer
Mentorship Focus
Supports learners across analytics, structured problem-solving, and applied ML foundations.

Ashutosh
Data Scientist
Mentorship Focus
Guides project work and practical implementation across core AI workflows.

Krishna
Applied AI Engineer
Mentorship Focus
Helps learners connect modern AI concepts with real engineering execution.

Vivek
Senior AI Engineer
Mentorship Focus
Focuses on helping learners build strong model-building intuition and debugging skills.
Generative AI Course Syllabus: Your 24-Week Learning Path
The syllabus is organised into four phases. Each prepares you for more advanced work, from understanding model behaviour to evaluating and integrating Generative AI applications. The public syllabus shows capability progression; detailed labs, datasets and solutions remain available only to enrolled learners.
Phase 1: Generative AI and Transformer Foundations
Weeks 1–6
Strengthen the foundations needed for the rest of the programme. Explore Python and machine-learning concepts, neural-network training, model evaluation, transformers, attention, tokenization, and introductory vision and language models. You'll begin building baseline experiments and investigating how model behaviour changes under different conditions.
- Python, mathematics and machine-learning refresher
- neural-network training and evaluation
- transformer, tokenization and attention behaviour
- vision and language model foundations
Output: Model evaluation workflow and documented transformer experiments.
Phase 2: Language, Vision and Multimodal Generative AI
Weeks 7–12
Explore how different generative model families are used. Study large language model behaviour, vision transformers, vision-language models, diffusion foundations, prompting strategies, and structured outputs. You'll examine how language and visual information can be combined within an application.
- generative model families and appropriate use cases
- large language model behaviour and limitations
- vision transformers, vision-language models and diffusion foundations
- controlled prompting and structured outputs
Output: Evaluated language or multimodal workflow.
Phase 3: RAG, Embeddings and LLM Fine-Tuning
Weeks 13–18
Learn two important approaches to improving Generative AI applications: supplying better information and adapting model behaviour. Work with embeddings, vector databases, hybrid retrieval, reranking, and source attribution while developing RAG systems. Explore LoRA and QLoRA fine-tuning, baseline comparisons, and the trade-offs between prompting, retrieval, and model adaptation.
- multimodal document workflows
- LoRA and QLoRA adaptation
- prompting vs RAG vs fine-tuning decisions
- chunking, embeddings, vector and hybrid retrieval
- reranking, citations and RAG evaluation
Output: Tested RAG system, fine-tuning comparison, and evaluation results.
Phase 4: Evaluation, Serving and GenAI Capstone
Weeks 19–24
Connect application integration, evaluation, and operational considerations. Explore model-serving fundamentals, quantization constraints, API packaging, guardrails, and controlled tool-using workflows. Combine appropriate components in the final capstone and assess the system's quality, limitations, and production-readiness gaps.
- quantization and inference constraints
- API serving and deployment packaging
- alignment concepts, evaluation and guardrails
- bounded tool use and agentic foundations
- capstone integration and readiness review
Output: Integrated GenAI application, architecture documentation, and evaluation evidence.
Design around the GenAI application
Once retrieval and model behavior are working, system design brings in capacity, service boundaries, permissions, failure recovery and operating cost.
See the AI System Design Course →Generative AI Tools, Hardware and Additional Costs
Tools are introduced for a defined engineering task: adapting models, building retrieval, evaluating outputs, processing multimodal inputs, exposing APIs or testing serving readiness. The curriculum is not presented as a collection of software badges.
Tools and Frameworks Taught
| Engineering job | Current tool families |
|---|---|
| Develop and adapt models | PyTorch, Hugging Face Transformers, PEFT and TRL |
| Build retrieval systems | LangChain or LlamaIndex, FAISS and a project-appropriate vector store |
| Evaluate RAG and model outputs | RAGAS, DeepEval and task-specific human-review rubrics |
| Build multimodal workflows | selected vision-language and diffusion model families |
| Expose and demonstrate systems | FastAPI plus Gradio or Streamlit |
| Package and test serving readiness | Docker and selected inference frameworks such as vLLM or TGI |
Laptop and Hardware Requirements
You need a laptop that can run Python development environments, Jupyter notebooks and Docker containers. A GPU is not required on your local machine — model training and fine-tuning labs use cloud-based GPU access. Confirm current hardware recommendations with admissions if you are unsure.
Included Compute and API Access
Model labs use open-weight models appropriate to the available compute environment. Larger models may be analysed or accessed through APIs. Confirm with admissions what cloud credits or API access are included in the current cohort, as these may change between batches.
Additional Paid Requirements
Cloud GPU usage, paid API subscriptions (such as OpenAI or Anthropic) and any third-party tool subscriptions needed for your own project work are not included in the course fee. Budget separately for these if you do not have existing cloud credits. The course does not require every listed product as a paid subscription — open-source alternatives are used wherever practical.
Generative AI Course Fees, Duration and What's Included
The School of Core AI Generative AI course runs for 24 weeks, approximately six months, through live online instruction. The published programme fee is ₹64,999. The course includes instructor-led training, recordings, six guided projects, technical reviews, assessment toward a School of Core AI completion certificate, and the stated career-support services.
- Duration
- 24 weeks / 6 months
- Format
- Live online
- Projects
- 6 guided systems
- Certificate
- Included
₹64,999
Confirm applicable taxes and payment terms before enrolling. Personal cloud GPU usage and paid API/tool subscriptions may cost extra.
Generative AI Course Certificate and Career Support
Completing technical training is about what you can demonstrate, not only the topics you've attended. Learners who meet the required coursework, project, and capstone assessment criteria receive a School of Core AI Generative AI Course Completion Certificate. The certificate reflects completion of the programme's requirements. It is not a university degree or a guarantee of employment.
Generative AI Course Certificate and Assessment
Learners who complete the required coursework and project criteria receive a School of Core AI course completion certificate. It represents completion of the program requirements; it is not presented as a university degree, government licence or guaranteed-employment credential.
Award criteria
Projects completed and evaluated
Assessment requirements met
Capstone system demonstrated
Placement and Career Support Included
Project portfolio presentation review
Resume and LinkedIn positioning guidance
Technical interview preparation
Opportunity sharing when available
School of Core AI does not guarantee a job, interview, salary, employer or placement deadline.
Does This Generative AI Course Include Placement Support?
Yes. SCAI provides career and placement support, which can include portfolio presentation, resume and LinkedIn guidance, technical interview preparation, and sharing relevant opportunities when available.
The objective is to help you present your skills clearly and prepare for work involving Generative AI applications, RAG systems, and LLM engineering. Placement support does not guarantee a job, interview, employer, salary, or hiring timeline.
AI Developer vs Generative AI vs LLMOps: Which Course Should You Choose?
Choose AI Developer when your main goal is to build AI applications and backend workflows. Choose the Generative AI specialization when you need deeper RAG quality, model adaptation, multimodal systems and evaluation. Choose LLMOps when your focus is releases, observability, scaling, governance and production reliability.
| AI Developer | Generative AI | LLMOps | |
|---|---|---|---|
| Primary goal | Build apps | GenAI model depth | Operate systems |
| RAG | Applied | Core | Production |
| Fine-tuning | Limited | Core | Release/Ops |
| Multimodal | Applied | Core | Supporting |
| Evaluation | App-level | Model/system | Production |
| Serving | API/app | Model/app | Scale/infra |
→ Build AI applications
Generative AI
→ Go deeper into GenAI models, RAG, fine-tuning and evaluation
→ Operate, serve, observe and scale LLM systems
Frequently Asked Questions About the Generative AI Course
The questions people ask most before enrolling — course fit, prerequisites, syllabus depth, career outcomes, fees, certification and placement support.
Ready to Build Real Generative AI Applications?
If you want to move beyond experimenting with AI models and understand how Generative AI applications are built, evaluated, and improved, explore the syllabus and discuss your background with the SCAI team. We'll help you understand whether the programme fits your experience, what you'll be expected to complete, and the current training arrangements.
