Ground answers in evidence
Build retrieval systems that cite sources and measure whether the right context was found.
Founding cohort · Applications opening soon
One cohort · Eleven-week calendar · Two starting pointsBuild GenAI systems that survive contact with the real world.
Move beyond impressive demos. Build a grounded, evaluated and optimised GenAI product, then learn how to deploy and operate it with production constraints in view.
- Format
- Live, guided cohort
- Entry
- Diagnostic + matched route
- Outcome
- Production-minded capstone
Live expert teaching01Hands-on build labs02Weekly feedback03Demo-day capstone
The transformationBuild it. Test it.
Build it. Test it.
Prove it works.
This is not a tour of fashionable tools. Every week adds a tested capability to one end-to-end GenAI product and asks you to explain the decisions behind it.
Engineer useful agents
Connect models to tools and workflows with clear boundaries, failure handling and human oversight.
Optimise with data
Balance response quality, latency, memory and cost using repeatable evaluations.
Operate what you build
Deploy an observable service with security, responsible-AI checks and an improvement loop.
Is this for you?
For technical learners ready to build something solid.
You will be a good fit if you are…
- A software, data, cloud or platform engineer
- A solution architect or technical consultant
- A Masters student seeking substantial project experience
- A technically curious graduate comfortable learning through code
Direct-entry learners should have…
- Working Python fundamentals
- A basic understanding of APIs and Git
- Some exposure to ML concepts or application deployment
- Time to build and iterate between live sessions
Not ready for direct entry? The diagnostic can place you into a guided technical foundation bridge before week one.
Choose your starting point
One cohort. Two ways to join.
The full calendar runs for eleven weeks. Foundation Entry learners begin together in Week 1. Direct Entry learners join that same group in Week 4 when the eight-week core begins.
Route AFoundation entry
Build the technical baseline first
Choose Foundation Entry if you need Python, Git, API and ML foundations before tackling the production-focused core.
- Complete the three-week foundation bridge
- Build a starter application and complete a readiness review
- Continue into Weeks 4 to 11 with Direct Entry learners
Both routes finish together with the same capstone criteria
Route BDirect entry
Join when the core programme begins
Choose Direct Entry if you already have working Python, Git and API skills and understand core ML and LLM concepts.
- Complete the diagnostic before Week 4
- Attend a short cohort onboarding session
- Join Foundation Entry learners for Weeks 4 to 11
You do not attend or wait through Weeks 1 to 3
Eleven-week programmeTwo starting points.
Two starting points.
One finish line.
Foundation Entry completes Weeks 1 to 11. Direct Entry joins in Week 4 and completes Weeks 4 to 11. From Week 4, everyone learns, builds and presents as one cohort.
Weeks 1 to 3Foundation bridgeRoute A starts here
W01Python for AI applications+
Work with variables, functions, packages, data structures and a local development environment through practical AI examples.
You buildPython utility and working development environment
W02Git, APIs and JSON+
Use version control, collaborate through a repository, call an API and handle structured responses.
You buildVersion-controlled API integration
W03ML and LLM foundations+
Understand training, inference, prompts, tokens and evaluation while completing the practical readiness task.
You buildStarter LLM application and readiness submission
Weeks 4 to 11Developer coreRoute B joins here
W04Foundations and developer setup+
Bring both routes together, frame a valuable use case and establish the shared API and evaluation environment.
You buildBaseline LLM application and engineering brief
W05LLM application patterns+
Work with tokens, embeddings, structured outputs, context and model APIs without treating prompts as magic.
You buildReliable, testable GenAI API
W06RAG systems and retrieval evaluation+
Design chunking, indexing, retrieval, citation and evaluation for grounded answers over private data.
You buildEvidence-backed RAG assistant
W07Agentic workflows and tool use+
Build bounded workflows that call tools, manage state, handle failure and keep a human in control.
You buildTool-using agent workflow
W08Fine-tuning and model optimisation+
Compare prompting, retrieval, LoRA/QLoRA, quantisation and serving choices against quality, speed, memory and cost.
You buildModel optimisation decision report
W09Evaluation, safety and responsible AI+
Create scenario-based evaluations for correctness, robustness, privacy, bias, prompt attacks and operational risk.
You buildAutomated evaluation and red-team suite
W10Production deployment and LLMOps+
Containerise, deploy, observe and improve a GenAI service with practical CI/CD and production guardrails.
You buildObservable cloud deployment
W11Capstone hardening and demo day+
Turn the weekly builds into one coherent product, defend the trade-offs and present the evidence behind it.
You buildPortfolio-ready capstone and technical walkthrough
Your capstone
Proof that you can make the whole system work.
Choose a real domain and build a GenAI application that retrieves private knowledge, uses tools where they add value, and exposes its quality, risk, latency and cost trade-offs.
- Working application and source repository
- Evaluation suite with measurable acceptance criteria
- Deployment, monitoring and security approach
- Architecture and model-optimisation decision record
- Live demo and technical walkthrough
01Domain data
→02Retrieve
→03Model + tools
→04Evidence
Evaluate · optimise · observe
Your founding instructors
Three views of what makes AI work.
Technical delivery, product judgement and enterprise reality are taught together so your capstone is useful, credible and operable.
KR
01Krishna Reddy Kuchalakanti
Cloud, DevOps and production AI systemsArchitecture, security, delivery automation, observability and the practical realities of operating AI workloads in the cloud.
SN
02Soma Shekar Nandikonda
Product direction, validation and valueConnecting technical choices to customer problems, product strategy, validation evidence and an adoption story stakeholders can act on.
RC
03Ramakrishna Chilukuri Venkata
Business analysis and enterprise use casesTurning complex enterprise needs into clear workflows, requirements, acceptance criteria, controls and demonstrable outcomes.
What is included
A cohort designed for momentum.
Expert-led sessions
Clear concepts, worked examples and the decisions practitioners make in real systems.
Guided labs
Structured starter code and practical milestones that keep the capstone moving.
Office hours and reviews
Get unstuck, test your reasoning and improve the evidence behind your choices.
Peer learning
Compare approaches, review work and build alongside people solving similar problems.
Questions, answered
Before you apply.
Are Route A and Route B separate courses?+
No. They are two entry points into one cohort calendar. Route A Foundation Entry begins in Week 1. Route B Direct Entry joins in Week 4. Both groups study together from Week 4 and finish with the same capstone in Week 11.
Does Route B wait for the first three weeks?+
No. Route B has a later published start date. Learners complete the diagnostic and a short onboarding activity before joining the cohort in Week 4.
Can a fresher or Masters student join?+
Yes. Learners who need the technical foundations select Route A Foundation Entry and begin in Week 1. Learners who already meet the technical baseline select Route B Direct Entry and join in Week 4.
What happens if I need more preparation after the bridge?+
The readiness review identifies the specific gaps. Learners receive targeted practice and can join a later cohort once the build-ready baseline is confirmed. The bridge is designed as preparation, not as an automatic pass.
Do I need machine-learning experience?+
You do not need to be an ML researcher. The programme teaches the concepts needed to make sound application, evaluation, fine-tuning and optimisation choices.
Which cloud and model providers will we use?+
The core patterns are provider-neutral. Labs use a practical reference stack, with guidance for translating the work to the ecosystem relevant to you.
What will I have at the end?+
A deployed GenAI capstone, evaluation suite, architecture and optimisation record, and a technical walkthrough suitable for a portfolio or internal proposal.
Will students receive a certificate?+
A certificate of completion tied to participation and a reviewed capstone is proposed for the founding cohort. Final criteria will be confirmed before applications open.
When does the cohort start and what will it cost?+
Dates and founding-cohort pricing are being finalised. Join the list to receive the schedule, application details and early-access offer first.
Ready to build beyond the demo?
Make your next GenAI project the one that ships.
Join the founding-cohort list for dates, pricing and the application pack.
Join the founding cohort