Programme context
Both courses run at Chitkara Business School for undergraduate business cohorts, and both are part of the same university relationship rather than separate engagements. They are built for different audiences: the AI and machine-learning course is taken by BBA students in logistics and supply-chain and aviation management, and is deliberately conceptual and no-code; Generative AI for Business is taken by BBA FinTech students and includes a technical foundation block alongside the applied syllabus.
The learning outcomes described below are the outcomes the courses are designed to produce. They are not a record of what any cohort achieved — no completion, grade or outcome data is published here.
University programmeTwo courses, one institutionUndergraduate business cohorts
BBA (FinTech)Generative AI for Business
- School
- Chitkara Business School
- Session
- 2026–27
- Contact hours
- 30 theory + 28 practical = 58 listed contact hours
Six chapters of theory run alongside 14 labs, moving from what generative AI is, through prompt engineering and workplace writing, to visual content and retrieval-augmented generation.
Learning objectives
- Explain how generative AI creates content, distinguish it from traditional AI, and map the tool landscape to workplace output requirements
- Apply structured prompt engineering — task, context, constraints, output format, role and iterative refinement — to produce testable business outputs
- Apply generative AI to workplace writing and analyse drafts through a four-dimension review gate: accuracy, tone, compliance and context
- Create AI-generated visual assets and presentations, and evaluate them against professionalism, brand-consistency, IP, bias and disclosure checklists
- Explain retrieval-augmented generation and verify AI outputs through a claim–source–match–context–action routine before business use
Curriculum structure
- Ch 1 — Understanding generative AI: vocabulary, output types, the prompt–output loop, hallucination, and a task-type × output-stakes decision matrix
- Ch 2 — Popular GenAI tools: the tool landscape, conversational AI, search-connected AI and grounding, image generation, and a tool-selection flowchart
- Ch 3 — Prompt engineering basics: the four ingredients, clear and testable instructions, role prompting, iterative prompting, prompt patterns and the prompt audit
- Ch 4 — GenAI for workplace writing: the guided drafting loop, structured email prompts, summary fidelity, report scaffolding, compliant marketing copy and the four-dimension review gate
- Ch 5 — GenAI for images, design and presentations: diffusion basics, five-part image prompt anatomy, brand kits, deck generation, visual quality audit and disclosure ethics
- Ch 6 — RAG and AI accuracy: why models give wrong answers, retrieval plus generation, grounded versus ungrounded answers, and the limits of RAG
Representative practical activities
- Running a deliberately vague prompt across two models and comparing the context each one invents
- Building a structured prompt from a messy business scenario, then repairing a flawed output through targeted follow-up prompts
- PII-safe prompting with placeholders, against a prompt audit checklist
- Drafting workplace emails from a six-part structured prompt, then reviewing the drafts for accuracy, tone, compliance and context
- Checking a summary for the omission and the invented fact, and scaffolding a report with explicit verification placeholders
- Taking one brief to several marketing formats, then rewriting a non-compliant version
- Iterative single-variable refinement of image prompts, and a professionalism and risk triage of generated slides
- Uploading a policy document and comparing grounded against ungrounded answers, then running a five-step verification routine
What students produce
- A capstone generative-AI workflow portfolio presented end to end — the brief, the tool selection, the generation, the review gates and the verification trail
- Reviewed workplace-writing deliverables: structured emails, verified summaries and report scaffolds
- A mini campaign set produced against a brand checklist, and an audited presentation deck
Technical foundation modules
The handout also describes technical foundation modules delivered as content beyond the approved syllabus, through separately scheduled extra classes and lab sessions. These sit outside the 58 listed contact hours above.
- Python basics — syntax, control flow, core data structures, functions, file handling, hands-on in a notebook environment
- NumPy for numerical computing, and Pandas for data handling, including a data-cleaning exercise on a FinTech transactions dataset
- Machine learning basics — train/test split, regression and classification demos, evaluation metrics, overfitting intuition
- Deep learning basics — neurons, layers, activation functions, and embeddings as learned numeric meaning
- The Transformer architecture — tokenisation, embeddings, self-attention intuition, and why context limits and hallucination follow from the mechanism
- RAG internals — building a mini pipeline end to end: chunking, embedding generation, vector storage and similarity search, and grounded answers with citations
Agentic AI and multi-step AI workflows in FinTech operations, listed for advanced learners.
Assessment approach
Both courses are assessed on a mix of continuous practical work and examination. The AI and machine-learning course assesses through a case study and through the prompt portfolio and AI research report, alongside an end-term examination. Generative AI for Business assesses through continuous lab evaluations — a prompt-engineering practical and the capstone workflow portfolio — alongside sessional tests and an end-term examination.
Assessment components, their weightings and their scheduling are set by the department. One handout presents its weighting table explicitly as a sample for reference. Weightings and dates are therefore not reproduced here.