Live cohort · hands-on

Building Agentic AI Systems

Go from a single LLM call to autonomous, production-shaped agents. A practical, build-along course where you ship one real assistant and level it up module by module.

Format
Live · online
Modules
7 hands-on
Duration
~30 hours
Build
Sales Ops Assistant
sales-ops-assistant
1
LLM calls → Agent SDKprompting, tools, API-level agency
2
MCP + RAGreal data sources, grounded answers
3
Autonomy & patternsReAct, memory, HITL, security
4
Evaluationeval sets, trajectory scoring, judges
5
Prompt & pipeline optimisationDSPy · GEPA
6
LoRA fine-tuningTinker API, end to end
7
Best practices & wrap-upcost, observability, multi-agent
Syllabus

One assistant, refactored as you learn

You build a Sales Ops Assistant early on, then keep refactoring the same project module after module — so every concept lands against working code.

01

From LLM Calls to an Agent SDK

The foundation — what an agent actually is.

Start at the metal: what a model call really does, how prompting shapes it, and how tool use works at the API level. You'll wire all of it together into a bare-bones Sales Ops Assistant that becomes the spine of the whole course.

model callspromptingtool useAgent SDK
Project:  a working, minimal Sales Ops Assistant
02

Giving It Real Data — MCP & RAG

Closing the two gaps the first build leaves open.

So far the assistant fakes its world: deal and analytics data is hardcoded JSON, and the sales playbook is one long document. You'll fix both — connecting live systems through MCP, and grounding answers in the playbook with RAG — by refactoring the assistant you already built.

MCPtool serversRAGretrieval & grounding
Project:  the same assistant, now backed by real data
03

Autonomy & Design Patterns

Now that there's something worth being autonomous over.

With tools, MCPs and RAG in place, the assistant can finally make decisions. We go deep on the patterns that make agents reliable rather than lucky — ReAct, CodeAct, Plan-and-Execute, Reflection and recursive language models — and when to reach for each. Along the way we cover the things autonomy forces you to handle: memory & state across steps and sessions, human-in-the-loop approval gates, and the security concerns that come with a connected, acting agent (prompt injection, sandboxing tools, least-privilege access).

ReActCodeActPlan-and-ExecuteReflectionRLMmemory & statehuman-in-the-loopagent security
Project:  an assistant that plans, acts, remembers and self-corrects
04

Evaluation — Knowing It Actually Works

You can't optimise what you can't measure.

Agents are non-deterministic and multi-step, so they fail in ways unit tests never catch. Before optimising anything, you learn to judge quality properly: building eval sets from real failures, scoring whole trajectories rather than just final answers, using LLM-as-judge (and avoiding its traps), and setting up regression tests so improvements don't quietly break something else.

eval setstrajectory scoringLLM-as-judgeregression testingoffline vs online
Project:  an evaluation harness for your assistant
05

Prompt & Pipeline Optimisation

More reliability from the same model — without hand-tuning.

Now that you can measure quality, you can improve it systematically. Stop tweaking prompts by hand: optimise prompts and whole pipelines programmatically against the metrics you built in the previous module, so the agent gets measurably better while the underlying model stays the same.

DSPyGEPAmetric-driven
Project:  an auto-optimised assistant pipeline
06

LoRA Fine-Tuning with Tinker

Adapt a model to your own domain.

Take it the last mile with lightweight LoRA fine-tuning, using the Tinker API end to end — so the model speaks your domain's language rather than relying on prompting alone.

LoRAfine-tuningTinker API
Project:  a domain-tuned model powering your assistant
07

Best Practices & Bringing It All Together

From a working agent to one you'd put in production.

We close by stitching every piece into one coherent system and covering what production actually demands: cost, latency and token budgets; observability and tracing for multi-step runs; reliability patterns like retries, fallbacks and timeouts; and how multiple specialised agents coordinate. You leave with a checklist and a finished assistant you understand end to end.

cost & latencyobservabilityreliabilitymulti-agent orchestration
Project:  your assistant, production-shaped and reviewed

Prerequisites

Basic knowledge of software design and architecture, and comfort with JavaScript. No prior ML experience required.

July cohorts

Pick the schedule that fits your week

Two ways to take the same 30-hour course — spread across weekday evenings, or condensed into focused weekends.

● Weekday evenings

Evening Cohort

10 evenings · Mon 6 – Fri 17 July 2026 (two weeks)
6:30 – 9:30 PM IST · 3 hrs / evening
10 sessions · 30 hours total
Live online · recordings included
Limited seats · small cohortJoin now
● Weekend intensive

Weekend Cohort

4 full days · Sat 12 – Sun 13 & Sat 19 – Sun 20 July 2026
10:00 AM – 5:30 PM IST · full days
4 days · 30 hours total
Live online · recordings included
Limited seats · small cohortJoin now

Ship your first real agent this July

Both cohorts cover the same syllabus and the same hands-on build. Pick a date and reserve your seat.

Join a cohort