PROJECT 01
Invoice extraction pipeline
PDF invoices go in, validated rows land in a Delta or Snowflake table, with failures flagged for review.
Claude APIJSON schemaDelta or Snowflake
01 / OVERVIEW
What this course is about
Generative AI is changing what data engineers build. In this course you call LLMs from Python, turn PDFs and emails into clean tables, build retrieval augmented generation over your own documents, and create agents that can safely query a warehouse. I hold the Generative AI for Data Engineers certificate, and the AI assistant on this website runs on the same ideas.
You learn by doing: every week has hands-on exercises on Google Classroom, and every module ends with something you built yourself.
01
Call LLMs from code
APIs, system prompts, structured output, cost and latency.
02
Extract data from documents
Turn PDFs, emails and free text into validated tables.
03
Build RAG systems
Chunking, embeddings, vector search and evaluation.
04
Create AI agents
Tool use, MCP and agents that query data safely.
05
Use AI inside data platforms
Snowflake Cortex, Databricks AI functions and Fabric Copilot.
06
Run AI in production
Guardrails, monitoring, PII handling and governance.
02 / CURRICULUM
Week by week
A clear plan from the first session to the last. The pace adapts to you: if you already know a topic we move faster, and if something needs more time we take it.
WEEK 1LLM foundations for engineers5 topics
- How LLMs and embeddings work
- Tokens, context and cost
- Claude API and Azure OpenAI from Python
- System prompts and structured output
- Retries, rate limits and logging
WEEK 2AI in the data stack5 topics
- AI coding assistants for SQL and PySpark
- Claude Code and Copilot workflows
- Snowflake Cortex and Databricks AI functions
- Document and PDF extraction pipelines
- Validating AI output with schemas
WEEK 3RAG and vector search5 topics
- Chunking strategies
- Embeddings and vector stores
- Databricks Vector Search and Cortex Search
- Building a docs Q&A
- Measuring answer quality
WEEK 4AI agents and production5 topics
- Tool use and function calling
- Model Context Protocol (MCP)
- A read-only text to SQL agent
- Guardrails, PII and governance
- Capstone demo and interview talking points
03 / PROJECTS
What you will build
Real projects you can put on GitHub and explain with confidence in interviews.
PROJECT 02
RAG over data documentation
A chatbot that answers questions about your tables and pipelines using your own docs, with sources.
EmbeddingsVector searchEvaluation
PROJECT 03
Text to SQL agent
An agent that answers business questions by writing SQL against a read-only warehouse, with checks before it runs.
Tool useMCPGuardrails
04 / TOOLS
Tools and technologies
The same tools used by data teams in real companies.
PythonClaude APIClaude CodeAzure OpenAISnowflake CortexDatabricksVector databasesMCPpandas
05 / IS IT FOR YOU
Who this course is for
Who it is for
- Data engineers who want to build AI features
- Python and SQL developers curious about LLMs
- Analytics teams adding AI to their platforms
- Engineers preparing for AI-focused data roles
What you need
- Python basics (functions, lists, dictionaries)
- SQL basics
- An API key for Claude or Azure OpenAI, usually a few dollars of usage
Roles it leads to
- AI Data Engineer
- Data Engineer
- Analytics Engineer with AI
- ML and AI Platform Engineer
06 / HOW YOU LEARN
What you get
Live and one-to-one
Every session is just you and me, never a batch. No recordings, so we can talk openly about your code.
3 sessions a week
Each session is 1 hour 15 minutes, on weekdays, weekends or both. We agree the schedule before starting.
Google Classroom
Assignments, projects, quizzes, interview questions and homework, all organised in one place.
Real projects
You build projects like the ones companies run, and get feedback on your code and design.
Interview preparation
Common questions, live coding practice and mock interviews with honest feedback.
Job placement support
CV, LinkedIn, applications and interviews. I stay with you until you are hired, as long as you do the work.
07 / PRICING
Simple pricing
You pay for your time with me, not per course. All prices are in US dollars.
Estimated total for Generative AI for Data Engineers at the typical pace (4 weeks)
About 200 USD
Full refund if it is not right for you after the first session. Go faster and it costs less.About 200 USD
08 / YOUR MENTOR
Hikmat Ullah
Senior Data Engineer at Algo · Lahore, Pakistan
I build data platforms with Snowflake, Azure Databricks, Azure Data Factory and Microsoft Fabric every day, and I have been mentoring since 2021. I teach what I use at work, and I stay with you until you land the job.
7+years in data
100+students trained
80+placed in jobs
1:1always, never a batch
09 / FAQ
Questions about Generative AI for Data Engineers
Do I need machine learning knowledge?+
No. This course is about using LLMs as building blocks. Python and SQL basics are enough.
How much will the API usage cost?+
For the course projects it is usually only a few dollars. I show you how to set limits and pick cheaper models for testing.
Will I learn Claude specifically?+
Yes. We use the Claude API and Claude Code a lot, and also compare with Azure OpenAI so you can work with whatever your company uses.
How much does it cost?+
200 USD per month whichever course you choose. You can also pay 100 USD every 15 days or 50 USD per week. All prices are in US dollars wherever you live.
What is the refund policy?+
If the training is not right for you after the first session, you get a full refund. After that, payments already used for completed sessions are not refunded, but unused prepaid time can be refunded.
Is there a job guarantee?+
Yes. As long as you attend your sessions and complete the assignments and projects, I keep helping you with CV, LinkedIn, mock interviews and job applications until you are hired.
Can I join from another country or time zone?+
Yes. Students join from many countries. We pick session times that work for your time zone before we start.
10 / KEEP GOING
Related courses
4 weeks
Prompt Engineering
Prompting patterns for engineers, analysts and AI agents, with real data tasks.
View course → 3 monthsPython Developer
Python for data work: pandas, APIs, automation, testing and real ETL.
View course → 3 monthsAzure Databricks Engineer
PySpark, Delta Lake, medallion architecture and Unity Catalog on Azure.
View course →START WHEN YOU ARE READY
Start the Generative AI for Data Engineers course
Tell me your goals and your time zone. We plan your schedule together, and your first session can start this week.