Train

03 — Developers & IT teams

Advanced AI training for developers and IT teams

The intensive engineering programme: development frameworks, model integration and APIs, custom solutions, performance and security.

For
Developers & IT teams
Modules
6
Delivery
On-site or remote
Length
Scoped per cohort
Request a briefHow these run

Why this programme

Teams rarely struggle to build a demo. They struggle where the demo meets real inputs, a cost ceiling and someone on-call — and that is exactly where most AI training stops.

This programme starts there. It draws on the multi-agent systems I architect in production, and every module ends with something running.

Curriculum

What it covers

  1. AI development frameworks

    The landscape without the hype: what LangChain, LangGraph and CrewAI are each for, and how to choose rather than collect.

    • LangChain
    • LangGraph
    • CrewAI

  2. Model integration & APIs

    Provider APIs, streaming, structured output, tool calling and the Model Context Protocol. The boundary work that turns a model into a feature.

    • APIs
    • MCP
    • Tool calling

  3. Retrieval and custom solutions

    Chunking, embeddings, retrieval strategy and evaluation — then agentic RAG for when a single lookup is not enough.

    • RAG
    • Embeddings
    • Evaluation

  4. Performance and cost optimisation

    Latency, caching, model selection and token economics. How to make a feature ten times cheaper without making it worse.

    • Latency
    • Cost
    • Caching

  5. Security and safe deployment

    Prompt injection, data leakage, access control and audit. The failure modes that do not exist in ordinary software.

    • Security
    • Prompt injection
    • Audit

  6. AI-assisted development

    Getting real throughput from Claude Code, Copilot and Cursor without accumulating review debt — from someone who rolled this out across an enterprise team.

    • Claude Code
    • Copilot
    • Code review

What the team can do afterwards

  • Ship a working AI feature against your own systems
  • Choose between a chain, an agent and a plain function — with reasons
  • Build retrieval you can evaluate rather than hope about
  • Cut latency and cost without degrading output
  • Recognise and defend against AI-specific attacks

Who it’s for

  • Engineering teams moving from AI demos to production
  • Platform teams setting internal standards
  • Tech leads reviewing AI code they did not write
  • IT teams integrating AI into existing systems

Prerequisites

  • Comfortable in Python or JavaScript
  • No prior LLM or ML experience required
  • A laptop and an API key for the session

Delivered personally by Viren Gajjar

Questions

About Advanced AI for IT Professionals

  • Do participants need machine learning experience?

    No. The programme assumes working knowledge of Python or JavaScript and builds from there. It is about engineering with models, not training them.

  • Can it run against our own codebase?

    Yes, and it works considerably better that way — integration and evaluation work becomes real rather than illustrative.

  • Is it tied to one model provider?

    No. Examples use whichever provider you already have access to, and the architecture discussion is deliberately provider-agnostic.

Ready to scope a cohort?

Tell me the team size, their starting point and what you need them to be able to do afterwards. You’ll get a proposed shape and a fixed quote back.

Request a briefAbout Viren