PRODUCT MANAGEMENT FOR AI & ML

Meet your instructor Han Vanholder. With a rich background in the tech industry, Han will introduce you to his craft and provide an overview of what you can expect to learn.

• Instructor introduction
• Course objectives & flow
• Brief overview, key moments, and rapid growth of the AI industry
• Q&A

What does an AI product manager actually do? Today we’ll break down the roles of the trade and the product lifecycle. Then we’ll take a look at where and how AI is used, exploring examples of unexpected places where AI delivers value today.

• Recent trends & the unreasonable efficiency of Deep Learning
• Product management skills for AI & ML
• Product management in the AI industry
• Extraordinary AI use cases

Let’s start with the basics. Decoding the complicated terms within AI is crucial to understanding how the field works. As you start to understand the value and cost drivers for AI/ML, you’ll learn how they integrate with the product life cycle.

• Fundamentals of AI & Machine Learning
• Models: the rocket engine of Deep Learning
• From training to deploying ML models
• AI lifecycle
• The relevance of new AI technologies for PMs

If your business has a problem, how can you pitch AI as the solution? Hone your ability to answer this question. Practice defining problems and target audiences to uncover the value AI/ML can bring them.

• Pick a problem & pitch AI as a solution
• Identify users & customers
• Develop & test the value hypothesis
• How to measure success & identify challenges

Assignment #1: Identify a work-related problem & develop an ML-based solution.

You’ve chosen a problem. You’ve defined your customers and users. Now you’ve got to create a deployment strategy. Let’s start by expanding on the selected problem and developing our product strategy, giving a deeper look into what an AI product manager does.

• Business needs, prioritization, and trade-offs
• MVP and MLV
• Do you need a developer strategy?
• Workshop: Developing a GTM strategy

Assignment #2: Develop a GTM strategy for your ML solution. (4 slides: target market, personas & developers, positioning & value proposition, and pricing strategy)

Data empowers and accelerates AI. In today’s lesson, you’ll get to practice checking the quality of a real data set. The tasks, such as having to find data and making sure it’s unbiased, are obvious. The execution, on the other hand, is challenging.

• Larger models, large training data sets
• Acquiring data & recognizing bias
• Getting to labeled data
• Mitigating poor data quality
• Workshop: Examine digits data set

Assignment #3: Select relevant data & data sets for your ML solution and outline a data strategy.

Over time computer vision has evolved. Cultivate a deeper understanding of its current limitations and demystify various models to ease your transition into AI/ML product management. It’s time to jump into its core principles and applications!

• Computer vision applications
• The historical evolution of computer vision
• Generative AI & popular models
• Training a model from scratch vs. using a pre-trained model
• Workshop: Build and train a digit classifier

Ramp up your AI knowledge. As we discuss foundational technologies, we’ll slide into the importance of large language models and the reason why they work. The principles, applications, and current limitations of sequential models will be tackled today.

• RNN, LSTM, BERT
• Transformers
• Large language models & multi-modal models
• Case Study: Facebook & applying LLMs to biotechnology 

Assignment #4: Refine your ML solution and design the high-level architecture in three slides: solution architecture diagram, key components, and reasons for selecting those components.

Besides counting dollars, there are other ways to gauge the success of your product. Let’s learn them! You’ll get into creating actionable steps based on metrics, deciphering which metrics are applicable, and making a feedback loop to keep improving your product.

• Importance of defining & measuring outcomes
• Choosing & defining metrics and KPIs
• Common measurement pitfalls
• Making measurements actionable
• Case Study: Code Copilot

Assignment #5: Define outcomes, key KPIs & measurement strategy for your ML solution. Articulate how these integrate into the solution and deployment architecture.

As a PM you need to ensure fairness and use AI responsibly. Buckle up because we’re going to launch ourselves into mitigating bias, giving users transparency, and accessing frameworks that minimize harm and maximize positive results. Champion ethics, and win user trust!

• Privacy: balancing innovation with data protection
• Robustness: safety & security in AI systems
• Governance & sustainability
• Tools to address issues
• Responsible AI strategy framework to minimize harm

Assignment #6: Identify ethical aspects of your ML use case & create a strategy to address them.

AI product managers have to juggle. You have marketing teams, developers, and clients interacting with your products on different levels. Plus, you have to collaborate with them all at once. Let’s discuss the path to deploying a product and ensuring it continuously reaches business goals.

• Stakeholder management: engineers, sales teams & customers
• Evolving your solution & backwards compatibility
• Case study: TF2.0
• Deployment management & post-launch monitoring

You’ve learned the little details to release a product, now it’s time to zoom out and look at the AI landscape. In this class, you’ll use AI to create solutions, be more productive, and generate future opportunities.

• Technology trends
• Graph Neural Networks
• Use cases
• Ads ranking, recommendations & personalization
• Productivity tools: harnessing AI to be more productive

Assignment #7: Identify three AI tools for your daily work, and describe their applications and improvements in one paragraph. Specify one metric for each to measure effectiveness.

Congrats! You’re about to reach the next level in your career, but first…interviews! Prepare your questions before class and get ready to meet our final guest speaker. After today, you’ll be equipped with industry insights on creating a top-performing resume and ways to market yourself as an AI/ML product manager.

• Exploring hard skills & qualifications
• Pros & cons of qualifications
• Career growth tips on presenting yourself (resume)
• Developing your interview skills
• Guest speaker Q&A panel (prepared questions)