Generative AI frameworks, APIs, models, and tools are being released faster than even the most astute data professional can keep up. After some discussion with members inside our private substack community, we’ve decided to host a 100 Days Of Generative AI challenge. With this free collaborative challenge, we’re providing a generative AI learning path for you to use to get up-to-speed on AI engineering, prompt engineering, and generative AI, in general.
To be clear, this challenge is meant to be completely collaborative. We are here to support and encourage one another in the process of getting upskilled and functional as builders of generative AI applications. You are welcome and encouraged to join, participate, and contribute in any way that you can.
I’m Lillian Pierson, the founder of the Data-Mania blog here – and I’m building this curriculum plan out of the very limited amount of time I have to research generative AI requirements and resources. There are gaps in the generative AI learning path provided below. That’s because I simply haven’t had time to conduct deep research in those areas. If you know good resources to fill those gaps, or any gaps that you identify in the generative AI learning path below, then please leave your suggestions in the comments and I will go back and add those to the list later.
Caveats On The Generative AI Learning Path Shared Below…
Two things to note here:
- I am a data product manager, so I sourced these requirements from LinkedIn job listings for both AI engineers and data product managers. Some of the recommendations are from more traditional data requirements (ie; SQL, Tableau, and Python), nonetheless – those are still base requirements for effectively managing generative AI products, so I have included them.
- AI engineering is quickly bifurcating into 2 branches. ML engineering (this is not new) and AI engineering (this is an emerging sub-discipline). To keep the time-to-value low here, within this generative AI learning path, I’m only making recommendations for the AI engineering route. Please read this post if you want more information regarding the differences between these sub-disciplines.
One last thing you need to know about the generative AI learning path shared below… It’s under development. I’ll be updating it on a regular basis as we all work together in the challenge to support each other’s professional development and growth. Please check back often for changes.
You’re Invited To Join Our 100 Days Of Generative AI Challenge (optional)
The mission of the 100 Days Of Generative AI Challenge is simple: To create a robust, supportive, and collaborative community that’s dedicated to supporting fellow data professionals in the learning and building of generative AI products and features.
Taking the courses that are prescribed in the generative AI learning path below is a great start, but there’s much to be said for accountability and networking support that’s only available inside of communities like the one we’ve set up for this challenge.
The community for this challenge will be hosted over on LinkedIn. Guidelines for sharing your work and supporting others will be provided within that group itself. If you would like to join this 100 Days Of Generative AI Challenge, please join our substack below and you will be automatically emailed with information on how to join the free LinkedIn community. (if you’re already part of our substack community, the details for joining this group were already sent to you in a past email titled: Join The Free 100 Days Of Generative AI Challenge)
And without further ado, let’s take a look at the generative AI learning path recommendations [LAST UPDATED Sept 8, 2023]
The Free Generative AI Learning Path
Be sure to start by reading the following articles:
After you’ve read those and developed a broad understanding of the requirements space for generative AI, next it’s time to start working through the generative AI learning path recommendations. Here are the recommendations I’ve come up with so far from my networking and independent research efforts.
Learning Path Legend
Each course recommendation is marked with:
- A time estimate – estimated hours until completion, and
- A color token to indicate the difficulty of technical pre-requisites.
Pre-Requisite Color Token Legend
🟢 = BEGINNER – Basic Python Only (if that…)
🟡 = INTERMEDIATE
REQUIREMENT: LLM APIs (Base Essentials)
LLM Foundation Model APIs:
- OpenAI
- Anthropic
- Cohere
Learning
Building
REQUIREMENT: LLM API Frameworks (Base Essentials)
LLM API Frameworks:
- Langchain
- LlamaIndex
Learning
- LangChain for LLM Application Development (5 Hours) 🟢
Building
- LangChain: Chat with Your Data (8 Hours) 🟢
- Example Project: A Hands-on Journey to a working LangChain LLM Application 🟡
REQUIREMENT: A grasp of AI, Large Language Models (LLMs), and prompt engineering, including Chain-of-Thought (CoT) prompting and Self-Consistency in CoT (Base Essentials)
Learning
- Generative AI with Large Language Models (AWS – 16 hours) 🟢 – This course is taught by my friends, Chris Fregly and Antje Barth.
Building
- Building Systems with the ChatGPT API (5 hours) 🟢
REQUIREMENT: Implementing Generative AI Solutions / Builds AI prototypes (Base Essentials)
Learning
- Generative AI with Large Language Models (AWS – 16 hours) 🟢 – This course is taught by my friends, Chris Fregly and Antje Barth.
Building
- How Business Thinkers Can Start Building AI Plugins With Semantic Kernel (5 hours) 🟢
- Building Generative AI Applications with Gradio (8 hours) 🟢
- AI For Good Specialization (? hours) 🟢
- Hugging Face NLP Course (64 hours) 🟡
REQUIREMENT: Python For Data Science / Jupyter Notebooks
Learning
- ChatGPT Prompt Engineering for Developers (OpenAI) (5 hours) 🟢
- LangChain for LLM Application Development (5 Hours) 🟢
Building
- Building Systems with the ChatGPT API (5 hours) 🟢
- Building Generative AI Applications with Gradio (8 hours) 🟢
- LangChain: Chat with Your Data (8 Hours) 🟢
REQUIREMENT: Tableau
Learning
- Tableau 2022 A-Z: Hands-On Tableau Training for Data Science (9 hours) 🟢 – This course is by my friend, Kirill Eremenko.
Building
😬 This is a resource gap – please leave a comment with a suggestion if you have a recommendation for a good resource that you have used yourself and that we can use to fill this gap.
REQUIREMENT: SQL (with certification)
Learning
- SQL for Data Science (14 hours) 🟢 – This course is by my friend, Sadie Lawrence.
Building
😬 This is a resource gap – please leave a comment with a suggestion if you have a recommendation for a good resource that you have used yourself and that we can use to fill this gap.
REQUIREMENT: AWS and Microsoft Azure Cloud + ETL Pipelines To Support Generative AI Products
Including AWS Glue (Cloud ETL)
Learning
- AWS Cloud Skill Support: skillbuilder.aws 🟢
- AWS Innovate Online Conference – AWS provides plenty of free training and support on this topic within these regular conference events.
Building
Additional resources
What would you add to this generative AI learning path?
You can help a lot of people by making suggestions and providing feedback on the generative AI learning path that I’ve shared here. Please share your tips on how we can improve it by submitting a comment on this blog post.
And again, if you want to participate with us in the free accountability community then please drop your dets in the form below and you’ll get an email with the details you need to join.
I hope to see you in there with us!
Warmly,
Lillian Pierson
ABOUT ME:
I am Lillian Pierson. I have 18 years of experience launching and developing technology products and delivering strategic consulting services. Additionally, I’ve also managed the development and launch of dozens of e-learning products; Products that educate learners on how to apply data science, data strategy, and business strategy to increase profits for their companies. To date, the products I’ve managed have been consumed by ~2 million learners and have generated over $6M in revenue for my clients.
I have launched over 40 products globally, delivered in 4 different languages. My products & go-to-market strategies have supported organizations as large as Walmart, Amazon, Microsoft, Dell & the US Navy. In fact, over the last 10 years I’ve supported 10% of Fortune 100 companies. Industries I’ve supported include Software as a Service, education, ecommerce, media, technology consulting, government services, finance, environmental consulting, oil & gas, and banking.
Besides my extensive business background, I’m also an accomplished data scientist & engineer, having held licensure as a Professional Engineer since 2014.
