HiWeb AI Academy

Become an AI Engineer

A hands-on 6-week program where you build the real systems behind modern AI products — retrieval, agents, reasoning, and multimodal AI — and deploy a portfolio-ready capstone to production.

6 Weeks 6 Projects Live + Hands-On Instructor: Maysam Gamini $2,800
A chat application interface grounded by retrieval and connected to agent tools, models, and a production deployment target AI Engineer / Assistant You AI Engineer grounded 3 sources retrieved · vector search Foundation Models Retrieval & Tools Agents Production Deployment evaluated · monitored · shipped live in 6 weeks
Instructor

Meet Your Instructor

Maysam Gamini

Maysam Gamini

Software Engineering & SRE Leader

Maysam Gamini is a software engineering and SRE leader with extensive experience in software architecture, CI/CD, performance engineering, technical leadership, and mentoring engineering teams. His background spans enterprise systems, development workflows, architecture, and production reliability. He is also active in the Austin technology and AI community.

Software Architecture CI/CD & Performance Engineering Austin AI Community
View LinkedIn Profile

Learn directly from an engineer who has shipped and operated production systems.

Register for the Cohort
Curriculum

Six projects. One production-ready capstone.

Every project is a complete system you design, build, and demo — not a toy notebook exercise.

01

Build an LLM Playground

Open the black box. You'll build an interactive playground that exposes exactly how a modern language model turns a prompt into text — the same mental model you'll rely on for every project after this one.

  • LLM foundations
  • Tokenization
  • Transformers
  • Model families
  • Pre-training
  • Post-training (SFT, RLHF)
  • Evaluation
  • Text generation
Hugging Face PyTorch BPE Tokenization Logit Sampling
You'll ship: a working LLM playground you can demo and extend.
A prompt flowing through a tokenizer, a stack of transformer layers, a sampling step, and output text Prompt Tokenizer Transformer Layers × N self-attention · feed-forward Sampling temp · top-p Output Text
02

Build a Production RAG Assistant

Ground an LLM in your own data. You'll build a retrieval pipeline that parses real documents, retrieves the right context, and generates grounded answers you can actually trust — and measure.

  • Prompt engineering
  • Document parsing
  • Chunking
  • Embeddings
  • Vector search
  • Retrieval & reranking
  • Grounded generation
  • RAG evaluation
pgvector Cosine Similarity Cross-Encoder Reranking Groundedness Scoring
You'll ship: a production-style RAG assistant grounded in real documents.
Documents flowing through chunking, embeddings, and a vector database, then retrieval into an LLM to produce an answer Documents Chunking Embeddings Vector DB Retrieval LLM Answer
03

Build an Ask-the-Web AI Agent

Give the model hands. You'll design an agent that plans, calls tools, and iterates — searching the live web, calling APIs, and reasoning over what it finds until it can answer confidently.

  • Agent architecture
  • Tool calling
  • MCP
  • Routing & workflows
  • Planning
  • ReAct-style loops
  • Orchestration
  • Agent evaluation
Model Context Protocol Function Calling ReAct Loop Tool Routing
You'll ship: an autonomous research agent connected to real tools.
A central agent node connected to web search, APIs, and MCP tools, with a loop indicating iterative planning Agent plan · act · observe Web Search External APIs MCP Tools Planner
04

Build a Deep Research System

One query, many angles. You'll build a system that runs parallel research threads, verifies what it finds against sources, and synthesizes a single well-reasoned report — inference-time reasoning applied to a real task.

  • Reasoning models
  • Web search
  • Parallel research
  • Source collection
  • Verification
  • Synthesis
  • Inference-time reasoning
  • Self-refinement
Reasoning Models Parallel Tool Calls Source Citation Self-Critique
You'll ship: a deep research workflow that produces verified, synthesized reports.
A query branching into three parallel research threads, converging into verification, synthesis, and a report Query Research A Research B Research C Verification Synthesis → Report
05

Build a Multimodal AI System

Move beyond text. You'll build a pipeline that reasons across text, image, and video — combining generation, understanding, and evaluation into a single multimodal content system.

  • Text, image & video
  • Image generation
  • Diffusion concepts
  • Multimodal models
  • Media pipelines
  • Multimodal evaluation
Diffusion Models Vision Transformers Cross-Modal Grounding
You'll ship: a multimodal content pipeline spanning text, image, and video.
A central multimodal model branching into text, image, and video outputs Multimodal Model Text Image Video
06

Capstone — Ship a Production-Ready AI Product

Bring it all together. Choose your own idea, architect it, build it, test it, deploy it, and demo it live — a complete AI product you design end-to-end and can show to employers or clients.

  • Choose
  • Build
  • Test
  • Iterate
  • Deploy
  • Demo
System Design Evaluation Harness CI/CD Deployment Live Demo
You'll ship: a portfolio-ready AI product, live-demoed to the cohort.
Six ascending steps — choose, build, test, iterate, deploy, demo — leading up to a launch point Choose Build Test Iterate Deploy Demo

Six real systems. One capstone you can put in front of a hiring manager.

Register for the Cohort
Fit Check

Is This Course for You?

You want to move from AI tutorials to building real systems.

You understand some AI concepts but want a structured, project-based path.

You want to build real RAG systems and AI agents, not just prompt a chat window.

You want hands-on experience with production AI workflows, not just theory.

You want portfolio-ready AI projects you can show, not just a certificate.

Highlights

Course Highlights

01 — Path

Structured AI Engineering Path

Six projects sequenced so each one builds directly on the last — foundations, then retrieval, then agents, then reasoning, then multimodal, then a capstone that ties it all together.

Six sequential steps rising left to right, each connected to the next 01 02 03 04 05 06
02 — Method

Visual System Explanations

Every concept is taught with a diagram first — how data actually moves through a RAG pipeline, an agent loop, or a training run — before you ever touch code.

A diagram being explained, shown as an architecture sketch with an annotation callout "why this shape?"
03 — Format

Project-Based Learning

No isolated exercises. Every module ends with a real, working system you built yourself — the kind of project that belongs in a portfolio, not a notebook.

A code window transforming into a shipped, checked-off project card Shipped
04 — Quality

Production-Oriented Code

You'll write code the way it needs to work outside a demo — with evaluation, error handling, and structure — not the shortest path to a screenshot.

A code module passing through a testing and evaluation gate before deployment Code Eval Suite Production
05 — Depth

Understand the Why Behind the Architecture

You won't just copy a pattern — you'll understand the tradeoffs behind chunking strategies, agent loops, and model choices, so you can make the right call on your own systems later.

A magnifying glass inspecting an architecture diagram, revealing the reasoning behind it

Not just what to build — the reasoning behind every architectural choice.

Register for the Cohort
Included

What You'll Get

Live & Interactive Sessions

Weekly live sessions with real-time Q&A — not pre-recorded lectures.

Hands-On Projects

Six real, working systems you build yourself — not guided copy-paste demos.

Course Materials & Recordings

Diagrams, code, and session recordings you can revisit anytime.

Peer Learning

Work alongside a cohort of builders tackling the same projects.

Instructor Feedback

Direct feedback on your project architecture and code from Maysam.

Certificate of Completion

A HiWeb AI Academy certificate recognizing your finished capstone.

Portfolio-Ready Capstone

A complete, demoed AI product designed end-to-end by you — ready to show employers or clients.

Everything you need to go from concept to a demoed AI product.

Register for the Cohort
Duration 6 Weeks
Format Live + Hands-On
Projects 6
Level Beginner → Intermediate
Instructor Maysam Gamini
Price $2,800
Register

Register for the Cohort

Save your seat — you'll receive a confirmation email with your personalized invoice.

What happens after you register

  1. Save your seat Your registration is saved immediately — no waiting on payment to hold your spot.
  2. Get your personalized invoice A confirmation email arrives with your invoice attached, itemized and ready to pay.
  3. Get course access Once confirmed, you'll receive cohort details ahead of the first live session.
Duration 6 Weeks
Price $2,800
FAQ

Frequently Asked Questions

What are the prerequisites?

Comfort reading and writing code in any language. No prior AI or machine learning experience is required — the course builds up from LLM foundations.

How much time should I expect each week?

Plan for one live session plus 3–5 hours of project work per week. The exact pace is up to you — the projects scale with the time you put in.

What happens if I miss a live session?

Every session is recorded, so you can catch up on your own schedule and still bring questions to the next live session.

Will sessions be recorded?

Yes — all live sessions are recorded and shared with the cohort, along with the diagrams and code covered in each one.

What will I build?

Six real systems: an LLM playground, a production RAG assistant, an ask-the-web agent, a deep research system, a multimodal AI pipeline, and a capstone product you design yourself.

Is this beginner friendly?

Yes, if you're comfortable writing code. The course is structured to take you from LLM fundamentals through to production systems — no prior AI background assumed.

Will I receive a certificate?

Yes — you'll receive a HiWeb AI Academy certificate of completion once you've finished and demoed your capstone project.

How does payment work?

After you register, you'll receive a confirmation email with a personalized invoice attached, along with Zelle and check payment instructions.

When do I receive course access?

Once your registration and payment are confirmed, you'll receive cohort details and access ahead of the first live session.

How do I contact the Academy?

Email academy@hiwebenterprise.com with any questions before or during the cohort.

Build AI Systems You Can Actually Ship.

Six weeks. Six real projects. One capstone you can put in front of a hiring manager — built with a live instructor, not a pre-recorded course.