The quality of an AI feature's output is largely determined by how the model is prompted. Product managers can apply different prompt optimization techniques to get more reliable outputs, reduce variance across user inputs, and build features that behave predictably at scale. Constructing the Prompt Every prompt arrives through one of two roles: a system message…
On September 21, 2026, Fadi Stephan presented Building High-Performing Teams at Agile Prague 2026. Below are some of the resources mentioned in the talk: Building High-Performing Team Slides High Performing Teams Canvas To learn more about how to build high-performing teams, check out the Building High Performing Teams Workshop.
The variety of available LLM models is valuable. More specialized options, more price points, more deployment flexibility. But that same variety is what makes selection difficult. Open-weight models like Llama and DeepSeek sit alongside commercial APIs from OpenAI, Anthropic, and Google. Domain-specific models trained for legal, medical, or financial work add another layer. Each comes with…
Human intelligence does three fundamental things. It perceives the world through the senses. It makes meaning from what it perceives. And it produces something in response, whether a decision, a piece of work, or a new idea. AI is advancing through these same three capabilities. Understanding where AI stands in each one is a more useful…
Prototyping is how you find out whether your business and design decisions hold before you build. With a conventional feature, two questions drive that process: do users want this, and can they use it? Every user gets the experience you designed. When the prototype reveals a problem, you can trace it to a decision your team…
In 2025, the Stanford AI Index recorded 362 documented LLM safety incidents, a 55% increase from the year before. These are not obscure edge cases. A court sanctioned a lawyer for submitting filings containing fabricated citations generated by an AI tool. In another case, Google's AI-generated search summaries confidently recommended adding glue to pizza and eating…
Building an AI feature involves design decisions that are easy to underestimate. Here are five patterns worth considering, and the factors that should shape how you apply them. 1. AI Input Design: Format, Scope, and Guidance The input pattern covers three distinct decisions: format, scope, and guidance. What modality the user interacts through. A text field,…
Most product teams designing AI features start with capability: what can this model do? That question shapes what you build, but three others shape whether it succeeds: What are users ready to accept? How much of the work should be automated? And where does a human need to stay in the picture? The Overton Window: Know…
Product teams have shipped dozens of features. They know how to scope, estimate, design, validate, and ship. So when an AI feature lands on the roadmap, they treat it like any other backlog item. That assumption is risky. AI features need to be treated differently. The seven areas where teams most often get this wrong are…
An AI PRD needs to account for behavior you can't fully predict and quality criteria that are subjective and evolve with real usage. That's because AI is fundamentally different from traditional software: it's probabilistic, not deterministic. That uncertainty runs through every section of the PRD: how you scope the work, what you require, how you measure…
AI products are measured differently from the products you've managed before. This glossary covers the metrics you need to evaluate performance, spot problems, and make decisions with your team. Not every metric applies to every product. Use this as a reference for the terms you encounter, not a list of everything you need to track. Many…
Building AI into a product is easy to get excited about and hard to get right. Most teams either start with the technology and work backward to find a use for it, or they identify real problems but generate ideas without any grounding in what AI can actually do today. Neither produces a roadmap worth building.…
The previous post in this series ends with a prioritized list of pain points your team has aligned on. What it does not tell you is whether those pain points are worth solving. These five questions give you a consistent way to stress-test any pain point before you commit resources to it. 1. Magnitude: How many…
Before you decide what to build with AI, you need to understand where it will actually make a difference. User experience mapping gives your team a picture of the user's existing reality: what they do today, where they hit friction, and which problems are worth solving. The output is a prioritized set of pain points your…
The cost of running an AI model is easy to measure. It shows up in cloud bills, API invoices, and unit economics spreadsheets, and that visibility makes it a natural anchor for pricing decisions. It shouldn't be. The Floor and the Ceiling For any AI product, you're working between two numbers. The floor is your full…