Masterclasses
Three hours on one hard thing, taught properly.
Focused AI masterclasses for engineering teams, conferences, and cohorts — RAG in production, context engineering, agentic design patterns, evaluation, and AI security.
- Teams
- Conferences
- Deep dive
- 2–4 hours
- Online or on-site
What this is
A masterclass is not a shortened course. It takes one problem your team is currently getting wrong, works through it live against real code, and leaves behind a decision framework. Most run two to four hours, with the room's own architecture on screen rather than a toy example.
Who it is for
- Engineering teams shipping their first AI feature
- Platform and data teams supporting AI workloads
- Conference and community organisers
- Product and engineering leadership
- Advanced student cohorts and research groups
Formats
Pick the shape that fits.
Team masterclass
3 hours · private
One topic, your codebase, your constraints. Capped at 25 so questions get answered properly.
Conference masterclass
Half day · 40–150 seats
A ticketed deep dive for your event, with a public repository attendees keep.
Public cohort masterclass
3 hours · open enrolment
Run with your community or ours, recorded and shared with everyone who registered.
Executive briefing
90 minutes · leadership
No code. What the technology can and cannot do, what it costs, and which bets are defensible.
Topics
Choose one, or ask for another.
RAG that survives production traffic
Context engineering: retrieval, ranking, compression, position
Agentic design patterns and when a plain loop wins
Model Context Protocol, end to end
Evaluation harnesses and LLM-as-judge, including its failure modes
Memory systems for agents that outlive a session
Prompt injection and LLM application security
Fine-tuning economics: when it pays and when it does not
Inference optimisation — quantisation, batching, KV caching
Choosing a model: capability, latency, and unit cost
Outcomes
What changes afterwards
- A shared vocabulary so architecture arguments stop going in circles
- One decision framework the team can apply the next day
- A working reference implementation, kept by attendees
- A written list of the specific risks in your current design
Deliverables
What you keep
- Reference repository for the topic
- Slides and annotated architecture diagrams
- Recording for internal reuse
- A short written follow-up on the questions raised in the room
FAQ
AI Masterclasses — questions
How long is a masterclass?
Two to four hours, with three the most common. Anything shorter turns into a talk; anything longer needs a lab structure, which is a workshop.
What is the difference between a masterclass and a workshop?
A masterclass goes deep on one topic and is led from the front with live code you follow along with. A workshop is hands-on and longer — participants build something themselves under supervision, and we debug their machines.
Can you use our own codebase or architecture?
Yes, for private team masterclasses. We take a walkthrough under NDA a week before, then teach against your actual retrieval pipeline or agent design rather than a generic example. This is where most of the value ends up.
How many people can attend?
Private team sessions are capped at 25 so discussion stays real. Conference masterclasses run up to 150. Public cohorts have no hard cap since interaction moves to a moderated Q&A channel.
Do attendees need to code along?
For technical topics, yes — bring a laptop with Python and an API key, and we send setup instructions three days ahead. The executive briefing has no code at all.
Can the masterclass be recorded?
Private sessions are recorded and the file is yours for internal use. Conference recordings are released on whatever terms you set with your speakers.
Do you run masterclasses online?
Yes, and most are. Online runs on your video platform with a shared repository and a live Q&A channel. On-site sessions are available for teams in India and for international events with travel arranged in advance.
Also available
01
Faculty Development Program
Hands-on AI faculty development programs (FDP) for universities and colleges — generative AI, RAG, and agentic systems, taught lab-first, online or on campus.
02
Online Courses on AI
Instructor-led delivery, course design, course development, or a full turnkey package — for AI programs in full stack AI, generative AI, agentic AI, and forward deployed engineering.
04
Workshops, Webinars and Seminars
Hands-on AI workshops, live webinars, campus seminars, and hackathon mentoring — from a 60-minute session to a two-day build.
Ready when
you are.
Send the audience, the headcount, and the window you are working with. You get scope and price in writing within two working days — no discovery call required first.
Enquire about masterclasses