Generative AI Foundation for Engineers
Most "intro to AI" courses stop at chatbots. This workshop teaches you text generation, multimodal vision, structured outputs, function calling, MCP, and reasoning models — with provider-agnostic Python code you can ship today.
What You Will Learn
- ✓Build a streaming chatbot using the same patterns behind ChatGPT
- ✓Extract structured data from messy text so your code can use LLM outputs
- ✓Give AI access to real tools via function calling
- ✓Analyze images and extract text with multimodal vision models
- ✓Apply chain-of-thought reasoning for complex problem solving
- ✓Ship 6 working AI projects you can demo to your team
Who Is This For
- Software engineers — who have used ChatGPT but never built AI features into their own apps
- Backend developers — who want to understand GenAI capabilities beyond simple text completion
- Technical leads — who need to evaluate which AI capabilities are practical for their products
Curriculum
6 modules, 23 lessons, ~6 hours
Text Generation & Chat
4 lessons
Multimodal AI
3 lessons
Structured Outputs
4 lessons
Function Calling & Tools
4 lessons
Model Context Protocol (MCP)
4 lessons
Reasoning Models
4 lessons
Frequently Asked Questions
- What makes this different from other AI courses?
- You build 6 real projects, not toy demos. Each module covers a different AI capability with provider-agnostic code that works with OpenAI, Anthropic, or Google.
- Do I need machine learning experience?
- No. This course focuses on using AI APIs to build applications, not training models. Basic Python is all you need.
- Which AI provider should I use?
- Any provider works. We use LiteLLM so you can switch between OpenAI, Anthropic, and Google AI Studio with one line of code. A free Google AI Studio key is enough.
- How long does it take?
- About 6 hours of hands-on work. Each module is self-contained, so you can complete them in any order after Module 1.
Generative AI Foundation Course for Beginners
6 hands-on projects. 6 core AI skills. Zero fluff.
Where engineers learn to build actual AI features
The AI API Maze. Sound familiar?
Chatbot Myopia
You know how to send a prompt and print a response, but you have no idea how to make the AI actually interact with your application's database.
The JSON Nightmare
You're asking the LLM to return JSON, and writing brittle regex to parse it because the model randomly adds markdown blocks and hallucinated keys.
Isolated AI
Your AI features feel bolted on. The LLM can't look up real-time information, search the web, or trigger actions in your system.
Framework Lock-in
You wrote your entire app using the OpenAI SDK, and now your boss wants to switch to Claude, requiring a massive rewrite.
If you only know how to generate text, you are missing 90% of what makes LLMs powerful in production.
The fix isn't better prompting.
It's mastering the 6 core AI capabilities.
Senior engineers aren't better because they know more syntax. They're better because they've:
Once you master these 6 capabilities, you can build almost any AI feature your product needs.
Zero fluff.
Just the absolute essentials.
We skip the theoretical history of neural networks. Within the first 15 minutes, you are streaming tokens. By the end, you are building a multimodal app that can see, reason, and execute code.
The Engineer's AI Toolkit
6 Hands-On Projects. 6 Core Skills.
Text & Streaming
Build a streaming chatbot using the exact same connection patterns that power ChatGPT.
Structured Data
Use Instructor and Pydantic to extract typed, validated data objects from messy text.
Agents & Tools
Implement Function Calling to give your AI access to APIs, databases, and the internet.
Vision & Reasoning
Build a multimodal app that extracts tables from screenshots and solves problems step-by-step.
6 Modules. 6 Core Capabilities.
Text Generation & Chat
Learn the fundamentals of AI text generation — from your first API call to streaming, context windows, system instructions, and temperature control.
Welcome & Setup
75 XPGet introduced to the workshop, understand LiteLLM, and set up your development environment
Your First AI Response
75 XPMake your first AI API call, understand the completion API, and learn about streaming
Context, Personality & Temperature
100 XPTeach AI to remember conversations, give it a personality, and control creativity vs consistency
Text Generation Mastery
100 XPApply text generation to real-world applications and test your knowledge
Multimodal AI
Teach AI to see — analyze images, extract text with OCR, read charts, and compare multiple images using vision models.
Teaching AI to See
100 XPUnderstand vision models, learn how images are sent to AI, and analyze your first image
OCR & Chart Analysis
100 XPExtract text from images with OCR and analyze charts and graphs with AI vision
Image Comparison & Applications
150 XPCompare multiple images, build real-world vision apps, and understand limitations
Structured Outputs
Transform unstructured text into reliable JSON data using JSON mode and Pydantic validation — the backbone of AI data pipelines.
From Text to Data
100 XPLearn why structured outputs matter and how to use JSON mode to extract reliable data from text
Pydantic Validation
100 XPAdd type safety to AI outputs using Pydantic models — BaseModel, type annotations, and Optional fields
Complex Nested Extraction
100 XPExtract deeply nested data structures like resumes, invoices, and email action items
Best Practices & Review
100 XPApply structured output patterns to real scenarios and test your knowledge
Function Calling & Tools
Give AI access to real tools — calculators, APIs, databases. Learn tool schemas, the agentic loop, and multi-tool agents.
Why AI Needs Tools
100 XPUnderstand why LLMs need external tools, learn tool schemas, and build your first function-calling example
The Agentic Loop
125 XPBuild the core pattern that powers all AI agents: call → check → execute → repeat
Multi-Tool Agents
125 XPCombine multiple tools into one agent — calculators, weather APIs, and product search
Tools in Practice
100 XPReview best practices for security, cost, and reliability with function calling
Model Context Protocol (MCP)
Learn MCP — the universal standard for connecting AI to external tools and services, like USB for AI.
Understanding MCP
100 XPUnderstand what MCP is, why it exists, and how it differs from direct function calling
Connecting to MCP Servers
125 XPConnect to a real MCP server, discover its tools, and understand the connection lifecycle
MCP Agent Workflow
100 XPBuild a complete MCP agent that discovers tools, asks AI to choose one, and executes it
MCP in Practice
75 XPUnderstand the MCP ecosystem, when to use MCP vs function calling, and review key concepts
Reasoning Models
Unlock step-by-step thinking with Chain-of-Thought prompting — solve math, debug code, analyze decisions, and build the course finale.
How Reasoning Models Think
100 XPUnderstand the difference between pattern matching and reasoning, and learn Chain-of-Thought prompting
Math & Logic Problems
125 XPApply reasoning to mathematical calculations and logical puzzles
Code Debugging & Decision Analysis
125 XPUse reasoning to debug code, review for errors, and analyze complex decisions
Course Finale
100 XPReview all 6 pillars, take the final assessment, and celebrate your AI Foundation mastery
Detailed Curriculum
See every system, every week, in detail before you decide.
Amateurs prompt. Engineers build capabilities.
APIs are easy.
Anyone can paste a prompt into the OpenAI playground and get a poem.
Engineering is hard:
- Guarantee the LLM returns valid schema-compliant JSON
- Route complex multi-step intentions to the correct internal tools
- Process images cost-effectively without blowing up token limits
- Maintain architecture when switching from GPT-4 to Claude 3.5
Stop writing toy scripts. Start engineering AI systems.
Learn From Someone Who Actually Does This
I am the Head of Engineering at Jobbatical (EU Tech), with 8+ years of leadership and 15+ years of total experience in the software industry.
"Most engineers are not blocked by ability, but by lack of real system ownership."
This accelerator exists to give you what most jobs never will.
Guest Sessions From Engineers at
Live sessions on System Design, Career Growth, and Interview Preparation.
FAQ
What makes this different from other AI courses?
You build 6 real projects, not toy demos. Each module covers a different AI capability with provider-agnostic code that works with OpenAI, Anthropic, or Google.
Do I need machine learning experience?
No. This course focuses on using AI APIs to build applications, not training models. Basic Python is all you need.
Which AI provider should I use?
Any provider works. We use LiteLLM so you can switch between OpenAI, Anthropic, and Google AI Studio with one line of code. A free Google AI Studio key is enough.
How long does it take?
About 6 hours of hands-on work. Each module is self-contained, so you can complete them in any order after Module 1.