Melillo-720x900718×900 42.7 KB

David Melillo

This book shows you how to integrate large language models (LLMs) and AI into your everyday data engineering workflows. It’s built on short, hands-on lessons that are designed to fit into a lunch break, so you don’t need to dedicate long blocks of time to upskill.

Here are some of the things you’ll learn:

• How to craft better prompts for AI to help with SQL and Python tasks like data cleaning, transformations, etc.

• Ways to use ChatGPT (or similar) to write, debug, and optimize your data code.

• Embedding AI into pipelines via APIs to automate repetitive tasks

• Working with messy, real-world data and extracting insights.

• Building “agentic workflows” (i.e. more autonomous components) to scale the expertise in your org.

Who it’s for

• Data engineers & architects who already know SQL & Python.

• Data professionals who want to boost productivity, especially on the tedious parts of pipelines.

• Folks curious about using LLMs beyond “just chatbots” — actually integrating them into production pipelines.

Why this might matter to you

Here are a few reasons I think this book could be a useful addition to your toolbox:

• “Lunch break scale.” The format means you can digest something new each day without carving out a full weekend. Great for steady progress.

• Tool-chain relevance. There’s a gap today: many engineers want to use AI for helping with the hairy bits (dirty data, edge cases, debugging), but the path isn’t always clear. This book seems to aim directly at that gap.

• Future-proofing. As more parts of data architecture get “augmented” by AI, knowing how to embed those tools well, cleanly, and with performance in mind is going to be a differentiator.

Things to watch out for

• Since AI & LLMs are moving fast, some tools/versions may evolve. Expect some parts to date quicker than more theory/architecture bits.

• Real-world constraints (cost, latency, security, governance) often complicate AI in pipelines. The “lunch-lesson” format is great, but applying at scale means needing to think about infra, error handling, monitoring, etc. Make sure you test in your domain.

• If you’re brand new to data engineering (SQL & Python basics, ETL/ELT patterns), you may need supplemental learning — this isn’t a beginner’s “intro to data engineering” book; it assumes some knowledge already.

• Full details: https://www.manning.com/books/learn-ai-data-engineering-in-a-month-of-lunches

Don’t forget you can get 45% off with your Devtalk discount! Just use the coupon code “devtalk.com” at checkout