Live cohort · 15 sessions

Build production-ready AI applications, from first principles to deployment.

Most AI courses stop at “here’s how to call an LLM.” This one covers the whole lifecycle — transformers, RAG, agent design patterns, evaluation, containers, CI/CD, and a real deployment — so you finish able to ship AI systems, not demos.

Express interest See the curriculum

15 sessions
5 modules
Hands-on project every week
Capstone + mock interviews

Who it’s for

Built for anyone with basic programming skills and the knack to build

You don’t need years of experience or a background in machine learning. If you can write a function, use a loop, and you’re curious enough to figure things out — you have what this cohort needs from you. We’ll teach you the rest.

Students

Get ahead of the curve with hands-on AI engineering skills that most degrees don’t teach.
Career switchers

Coming from a non-tech background but comfortable with programming basics? This is your on-ramp into AI.
Engineers on legacy tech

Working in older stacks or maintenance-heavy roles? Add the AI layer that keeps you relevant.
Mid-career professionals

Ready to move your career forward — bring your experience and pick up the AI engineering skills to match it.
Software Engineers

Add LLMs, retrieval, and agents to the systems you already build.
Data Scientists

Turn notebooks into deployed services with evaluation and monitoring.

Prerequisite: comfort with basic programming — variables, functions, loops, and running code on your own machine. That’s it. Everything else, including the Python and Git you’ll need, is covered in Module 1.

Curriculum

Five modules, in the order teams actually build

Foundations first, then how models work, then building with them, then the engineering that makes it survive production.

Module 1 · 3 sessions

AI & Software Foundations

The landscape, Python and Git for AI work, and an AI-assisted development workflow.

View sessions → Module 2 · 2 sessions

Understanding Large Language Models

Transformers end to end: attention, embeddings, tokenization, training, inference.

View sessions → Module 3 · 3 sessions

Building LLM Applications

Prompt and context engineering, RAG and hybrid retrieval, evaluation and guardrails.

View sessions → Module 4 · 4 sessions

AI Engineering

LangGraph orchestration, agentic design patterns, production practices, MCP and deployment.

View sessions → Module 5 · 3 sessions

Become Industry Ready

Team capstone, mock AI-engineer interviews, and sessions with industry practitioners.

View sessions →

Outcomes

What you’ll have built by the end

Every module ships something. You leave with a portfolio, not a certificate.

An AI-powered chatbot grounded in your own documents with RAG
A multi-agent workflow that plans, researches, and validates
A production REST API built with FastAPI
An AI application deployed and running on AWS
An evaluation dashboard with golden datasets and metrics
MCP-enabled tool integrations other clients can call
A production-ready GitHub portfolio you can put in front of hiring managers

The details

Applied AI Cohort & pricing

Small, live, and cohort-based. Sessions are recorded so you never lose a week.

Start date: To be announcedJoin the list to hear first
Duration: 15 weeksOne session per week
Format: Live onlineRecorded and available afterwards
Time commitment: 6–8 hrs / weekSession plus the weekly project
Cohort size: Limited seatsKept small for feedback and reviews
Price: To be announcedEarly-bird pricing for the first cohort

Instructor

Taught by someone who does this in production

Nikhil Shrimali

Lead AI Engineer · Applied AI, Guardian Life

Nikhil has spent 11+ years building and productionizing enterprise-scale ML and generative AI systems, with deep experience in LLM agents, prompt and context engineering, NLP, RAG, and rigorous evaluation of open-ended generative output.

He leads applied AI at Guardian Life, where he owns a flagship underwriting-automation initiative — running a cross-functional team and acting as the bridge between business and technical stakeholders. He architected the agentic system behind it, combining context engineering, human-in-the-loop validation, and continuous evaluation into something reliable, scalable, and production-ready. Before that, at American Express, he shipped LLM-based conversational AI on retrieval-augmented generation and engineered a Neo4j Graph RAG architecture for entity disambiguation and contextual reasoning.

The spine of this curriculum is his day job, not theory: agents and orchestration, RAG and Graph RAG, LLM-as-judge and BERTScore evaluation, FastAPI, Docker, Kubernetes, and CI/CD — plus the AI-assisted coding workflow (Claude Code, Cursor) he uses to build with daily, which you’ll pick up in Session 3.

11+ yrs: Building ML and AI systems in production
100+: Engineers trained
15+: Production AI systems deployed