2 partsFoundations, then production
14 weeksFri–Sun, 6–8 PM Central
1 capstoneProduction AI application
AWS or AzureChoose one cloud provider

The bootcamp

Two parts. One application. Fourteen weeks.

Part 1 teaches the foundations of AI engineering on your own laptop. Part 2 takes the same project to production in the cloud. Take them back to back, or start at Part 2 if you already ship Python services.

Part 1 · 7 weeks

Foundations

$1,000

Build a voice AI agent on your laptop: a streaming LLM service, a voice pipeline, search with sources, tools, your first multi-agent system, evaluation, safety tests, and a human handoff.

  • How LLMs work, and a service you can test
  • Real-time voice: speech in, speech out
  • Embeddings, chunking and search
  • Grounded answers with citations (RAG)
  • Agents, tools and a multi-agent system from scratch
  • Evaluation and AI security
  • Human handoff and shipping your work

For you if you can read basic Python and use a terminal. No cloud account needed.

Part 2 · 7 weeks

Production & LLMOps

$1,200

Deploy and run the agent for real: Terraform on AWS or Azure, access control across companies, order and recall tools with a knowledge graph, a multi-agent system with a guardrail agent, tracing, monitoring, and recovery.

  • Cloud foundation with Terraform and CI
  • A streaming app deployed with rollback
  • Searchable knowledge in Weaviate
  • Voice with grounded answers
  • Safe order lookup and human handoff
  • Recall tracing with a knowledge graph and a multi-agent system
  • Tracing, monitoring, deployment and recovery

For you if you finished Part 1, or you pass a short entry check. See Part 2 week by week.

The first class starts here

A business question needs more than a fluent answer.

In Week 1, students start with a realistic wholesale order-support case: a buyer needs to know whether damaged products can be replaced and whether a recall affects their shipment. The application must check the right records, explain its evidence, and know when to involve staff.

Warehouse worker inspecting packaged coffee products beside a tablet
A question students build for
“Two cases in order ORD-1042 arrived damaged. Can I get replacements, and was the lot recalled?”
  1. Policy
    What does the approved damage policy say?The RAG app retrieves the current passage from Weaviate and shows its source.
  2. Order
    What did this customer actually buy?A read-only PostgreSQL tool checks the signed-in buyer’s organization.
  3. Recall
    Did the recalled lot reach this order?A knowledge graph query follows the supplier lot through shipment and order relationships.

The outcome: a sourced answer, a clear “I don’t know,” or a safe handoff to staff.

What you build in Part 2

One production AI application, built stage by stage.

Start with a real business case and grow the same application each week. By the final class, you can demonstrate the code, cloud deployment, evaluation evidence, and operating guide.

01Week 2

Streaming LLM web app

Ask a question in the browser and watch the model answer appear. Handle slow calls, empty input, errors, and usage.

02Week 3

Searchable knowledge base

Turn approved documents into versioned passages. Compare keyword, vector, and hybrid search in Weaviate.

03Week 4

Agentic RAG assistant

Route a question to retrieval, answer from permitted evidence, show a citation, and abstain when sources cannot support a claim.

04Week 4

Voice experience

Let a buyer ask aloud and hear a grounded answer while keeping the existing text and retrieval paths working.

05Weeks 5–6

Secure business tools

Check only the buyer’s orders in PostgreSQL and trace recalled supplier lots through a knowledge graph.

06Week 7

Operated cloud system

Release tested images, inspect traces and alerts, recover from backup, and demonstrate the stack on AWS or Azure.

Your capstoneA working GitHub repository, deployed application, automated checks, evaluation report, monitoring dashboard, and operating guide. It is evidence you can open in an interview or a client meeting and walk through line by line.
AI application layerPython, FastAPI, LLM APIs, structured outputs, and agent tools.
Grounded knowledgeEmbeddings, vector search, RAG, Weaviate, and retrieval evaluation.
Business dataPostgreSQL, authorization, human handoff, knowledge graphs, and graph queries.
Realtime experienceBrowser audio, transcription, speech output, and voice interaction.
Cloud deliveryDocker Compose, GitHub Actions, Terraform, and AWS or Azure.
LLMOpsTracing, evaluations, metrics, dashboards, backup, and recovery.

How it works

The model speaks. Your code decides what it may know and do.

Students build an explicit decision path around the model. That makes answers inspectable and lets each data source do the job it is suited for.

1

Classify the question. Is it about policy, a specific order, a recall relationship, or a request for a person?

2

Fetch trusted evidence. Use approved documents, an authorized order lookup, or a graph path. A question may need more than one tool.

3

Compose and check the answer. Name the source, keep customer data scoped, and offer a handoff when evidence or permission is missing.

What you will learn

The skills behind an enterprise AI agent people can trust.

Each skill is practised on the same order-support application, so you see how retrieval, tools, safety checks, and operations depend on each other.

Retrieval you can measure

  • Evaluate a RAG system with expected sources and no-answer questions.
  • Compare keyword, dense-vector, and hybrid retrieval for the same business question.
  • Use synthetic or permission-cleared data to improve retrieval without exposing private records.

Agents that pick the right tool

  • Build a multi-agent system from scratch: an orchestrator that plans each turn and specialist agents that each own their tools.
  • Build an agent that can retrieve evidence, call permitted tools, or hand a case to staff.
  • Ground complex questions in documents, order facts, and knowledge-graph relationships.
  • Learn how an MCP server can expose a protected tool as an optional advanced extension.

Safety built into the workflow

  • Route policy, order, recall, and escalation requests through predictable decision paths.
  • Add safeguards so the agent cannot invent access rights, change a record, or reveal another organization’s data.
  • Evaluate tool choice, grounded answers, denied access, and human handoff behavior.

Operations from the first week

  • Package services with Docker and provision cloud infrastructure with Terraform.
  • Use GitHub Actions for automated checks and deployment releases.
  • Trace failures, monitor health, restore data, and document the operating process.

Who it’s for

For developers who want to build AI agents for the enterprise.

You already write Python. What you want is the engineering around the model: where its facts come from, what it is allowed to do, and how you know it still works next week.

Developers adding AI to a product

You build web or backend software and have been asked to add an assistant. You want it to answer from your company’s data and stay inside its permissions.

Data and ML practitioners who want to deploy

You can prototype in a notebook. You want to package, test, deploy, and monitor the same idea as a service other people rely on.

Founders and freelancers building for clients

You are scoping an AI product for your own company or a client. You want to know what a dependable version costs to build and run.

Newer to code? Start with Part 1: you need to read basic Python and use a terminal. Looking for a no-code course? This is not it. Talk with us first and we will tell you honestly.

Part 2 curriculum

Seven weeks, with something working every Sunday.

Friday introduces the problem and the engineering decisions. Saturday is a guided build. Sunday is for finishing the checkpoint, running the tests, and showing it for feedback. Students complete the assignment in their own repository before the next Friday.

21 live sessions · 7 assignments · 8–10 hrs/week
Foundation · Weeks 1–3
01

Welcome, real use case, and local foundation

Trace the buyer’s damaged-order and recall question, map where private data lives, and make the first application boot from a clean clone.

Friday Map the buyer journey, system boundaries, failure cases, and AWS or Azure path.

Saturday Build the Python API and browser page; run Docker Compose, a health test, CI, and a Terraform plan.

Build checkpoint

Working starter application

A reviewer can boot the stack, see the passing GitHub Actions check, and inspect the infrastructure plan.

  • Python
  • Docker Compose
  • GitHub Actions
  • Terraform
02

How LLMs work, then a streaming app

Learn tokens, inference, context, prompting, latency, usage, and why a model can sound certain without knowing a fact.

Friday Compare model responses with and without evidence; define when the app should answer or abstain.

Saturday Stream a server-side model response, test timeout behavior, publish a versioned image, and deploy it.

Build checkpoint

Text app on a cloud host

Show a streamed answer, a graceful failure, a passing CI run, and a rollback to an earlier image.

  • LLM API
  • Python
  • Docker
  • Terraform
  • AWS or Azure
03

Embeddings, Weaviate, and reliable search

Learn how documents become chunks and vectors, and why a close match is a candidate source rather than proof.

Friday Compare keyword, vector, and hybrid retrieval; define expected sources for ten test questions.

Saturday Ingest versioned documents, build a search view, test access filters, and update the cloud stack.

Build checkpoint

Searchable policy knowledge base

Find an exact passage and its version, then rebuild a deleted search index from the source documents.

  • Embeddings
  • Weaviate
  • Retrieval tests
  • Docker Compose
Agent and voice · Weeks 4–5
04

Grounded answers and a voice interface

Connect the reviewed Week 3 index to an agent decision path. A spoken question can retrieve policy and produce a short sourced answer.

Friday Add a browser audio room, short-lived access token, agent participant, transcription, and speech output.

Saturday Add the policy search tool, source cards, refusal and handoff logic, and a golden question set.

Build checkpoint

Text and voice RAG assistant

Demonstrate a cited policy answer and an unsupported question that produces an honest abstention.

  • Agent tool calling
  • Weaviate
  • Realtime audio
  • Evaluation
  • DSPy optional
05

Authorized order lookup and human handoff

Combine live order facts with approved policy while keeping one organization’s records invisible to another.

Friday Create Postgres tables and migrations; bind the signed-in buyer to read-only order lookup.

Saturday Route order and policy questions, draft a support case after confirmation, and hand uncertain cases to staff.

Build checkpoint

Safe order resolution

Show a valid order answer, a denied cross-organization lookup, an evaluation result, and a human handoff.

  • PostgreSQL
  • Authentication
  • Agent tools
  • Tenant safety tests
Graph and operations · Weeks 6–7
06

Investigate recalls with a knowledge graph and a multi-agent system

When a supplier recalls a lot, identify affected customer orders and show the exact shipment path. Then build a multi-agent system from scratch: an orchestrator hands the question to Research, Data and Graph agents, and a Guardrail agent checks the answer before it is spoken.

Friday Model supplier, lot, shipment, product, and order relationships in a knowledge graph.

Saturday Add a permission-checked recall tool and project the graph from PostgreSQL.

Sunday Split the agent into an orchestrator and specialist agents, add a guardrail agent, and test every hand-off.

Build checkpoint

Verifiable multi-agent recall investigation

Trace a seeded lot to affected orders through the specialist agents, and show the guardrail blocking an unsupported answer.

  • Multi-agent systems
  • Knowledge graphs
  • Graph queries
  • PostgreSQL
  • Weaviate
07

Trace, monitor, deploy, and recover

Find why an answer is slow or wrong, then demonstrate that another engineer can operate the complete system.

Friday Follow one request through transcription, the orchestrator, each specialist agent, the guardrail, model response, and speech.

Saturday Deploy pinned images, trigger an alert, restore a backup, rebuild derived data, and rerun the tests.

Build checkpoint

Observable cloud deployment

Show the live app, a trace, a dashboard, a safe failure path, and a completed restore runbook.

  • OpenTelemetry
  • Phoenix
  • Prometheus
  • Grafana
  • AWS or Azure

Final demonstration

Show the system working—and recovering.

Students finish with a repository, cloud deployment, evaluation report, and operating guide. The final review follows one buyer request through the whole stack.

  • Answer a policy question with a source and refuse an unsupported claim.
  • Check an authorized order and trace a recalled lot without exposing another organization’s data.
  • Show the trace and dashboard for a failure, then restore records and rebuild search and graph data.

Dates and times

The next cohort, part by part.

Three 2-hour evening sessions every week, Friday to Sunday, so you can keep your day job and plan your calendar before you enrol.

Next cohortDates announced soon
Friday
6:00–8:00 PM Central · concepts and design
Saturday
6:00–8:00 PM Central · guided build
Sunday
6:00–8:00 PM Central · finish, test and demo
Order
Part 1 runs first. Part 2 starts the week after Part 1 ends.
Your week
  • 6 hrsThree 2-hour live sessions: design Friday, build Saturday, finish and demo Sunday
  • 2–4 hrsFinish the assignment in your own repository and record evidence
  • 8–10 hrsTotal, with the heaviest weeks when you deploy (Weeks 2, 5, 6 and 7)

Tuition

Tuition for each part.

Part 1 — Foundations$1,000
Part 2 — Production & LLMOps$1,200

Payment agreement available if needed.

Talk with us about the class

Ask about the curriculum, request a call, or discuss a payment agreement.

Call: (945) 347-3454 Text: (945) 347-3454 WhatsApp: (945) 347-3454

Reserve a seat or ask a question

Share a few details and we’ll contact you within a day to talk through the class and next steps.

We store these details to contact you about the class. See our privacy page.

After the class

Launch your own app or prepare for an AI engineering role.

The final project gives you working software to show. If you want help with your next step, choose the path that fits your goal.

App path

Build your own AI application

Bring your own use case. We help you shape the product, design the experience, build and deploy an MVP, and make a go-to-market plan for reaching your first users.

Use case → design → build → deploy → launch
Career path

Prepare for the job search

Turn your course project into a portfolio case study. We help you improve your resume and GitHub, practice technical interviews, and explain the engineering decisions behind your build.

Portfolio → resume → interviews → demo

Live class format

Three evening sessions each week. One clear build target.

Every week combines an engineering concept with an application students can demonstrate. Plan for about eight to ten hours a week, including live sessions and independent work.

Two developers working side by side at monitors showing code
Friday

Understand and design

Review the user problem, draw the data flow, compare approaches, and decide what safe behavior looks like.

Saturday

Build and test together

Implement the week’s feature, run the stack, test a failure case, and see the expected checkpoint.

Sunday

Finish and demonstrate

Get help finishing the checkpoint, run the tests, and show a short demo for feedback. Wrap up the assignment in your own repository before Friday.

Delivery: live Friday, Saturday and Sunday, 6–8 PM CentralWeekly work: about 8–10 hoursDeployment: Docker Compose on AWS or Azure

Before you start

Bring coding skills. We build the AI system together.

You should be comfortable with basic Python, Git, and a terminal. You do not need to have built a production AI application before the class.

Your laptopRun Python and Docker Compose, and keep your work in a GitHub repository.
Model accessBring access to an LLM API. The current course plan uses GPT-5.4 mini for language responses and separate speech services for voice.
One cloud account (Part 2)Choose AWS or Azure for the Terraform and deployment labs. You deploy to one provider.
Budget awarenessModel calls and cloud resources can cost money. Set a spending alert before deployment and remove resources when finished.

Questions

Before you decide.

Something not covered here? Ask us directly.

What do I need to know before I start?

For Part 1, you need to read basic Python and use a terminal. For Part 2, you should be comfortable with Python, Git, and Docker. You do not need machine learning experience or to have shipped an AI application. If you are unsure, send us a message and we will look at it with you.

Do I have to take Part 1 first?

No. If you already build Python services with Git and Docker, you can start at Part 2 after a short entry check. If you are newer to code, start with Part 1.

How much time does it take each week?

About 8 to 10 hours: three live sessions of 2 hours, Friday to Sunday from 6:00 to 8:00 PM Central, plus 2 to 4 hours to finish the assignment. Weeks 2, 5, 6 and 7 include cloud deployment and run toward the top of that range.

What if I miss a live session?

Tell us before you enrol if you expect to miss sessions, and we will tell you how this cohort handles it.

What will I have at the end?

A GitHub repository, an application deployed on AWS or Azure, an evaluation report, a monitoring dashboard, and an operating guide, plus a recorded final demonstration.

What will I spend on models and cloud?

Model calls and a small cloud host cost money on your own accounts. We show the cost of one request in Week 2, set a spending alert before you deploy, and remove resources at the end.

Which cloud do I use?

You choose AWS or Azure and deploy to that one provider for the whole course.

Can I pay in instalments, and what is the refund policy?

A payment agreement is available if you need one; ask us before you enrol. Ask us for the refund terms before you pay.

Is the after-class app or career help included?

Ask us before you enrol and we will tell you what is included.

Ready to build enterprise AI agents?

Request a call or text us