Ryan Wigley

We (Tech) spent the last three years obsessing over how we interact with language models - chatbots, agents, AI-powered workflows. Meanwhile, a more fundamental shift has taken place: prompting is giving way to context engineering and how you interface matters very little. When you need a problem solved with AI, asking a chatbot out-of-the-blue might give you the occasional good result. But for exceptional results, models need information about you, your problem space, and your expectations. Do it well and every thread feels like chatting with a partner that’s just as deep-in-the-weeds as you are.

Evolution Below the Surface

It's ironic how we obsessed over the interface while the real breakthrough was happening in the background. It wasn't about how we talk to AI. It was about what the AI knows before we even start talking.

Since ChatGPT's breakout in 2021, the scope of context we can engineer into these systems has expanded in three distinct stages:

Stage 1: Prompt-only context. "Hey, create this file for me based on some examples. Here's good, what's bad." The AI only knows what you tell it in that moment. Success meant obsessing over every word in your prompt.

Read more »

You’ve built a successful newsletter, but now life’s getting busier after your 50th issue. Whether it’s a move, new baby, or sponsorship commitments, you’re finding it’s challenging to keep the quality up without burning out. This is where copilots come in — AI partners that transform how we tackle work. Give it the right base of knowledge and you can create a personalized thought partners that understands your context, style, and goals.

I'll share three specific copilots I've built in Claude Projects and how you can create your own - I’m a big believer they can improve your productivity and thinking.

Read more »

A PM’s practical guide to using AI: mental models, prompts, and tools.

22 minute read

These last 3 years have made me rethink how I approach problem-solving as a product manager (PM). AI has exploded, software tools are plentiful, and cloud resources are more accessible than ever. In the sections below I will lay out how I approach defining problems and solving them with AI. I’ll also provide some tips, tools, and prompts I use for my work.

Before diving into practical advice, let’s walk through how the software industry has evolved.

Read more »

I've been finding myself thinking a lot about young Ryan lately. That kid who would spend hours hunched over the family Mac, figuring out how to program his first game in HyperCard. I can still feel the rush of successfully animating that crude stick figure to move across the screen - I was convinced I'd just created the next Duke Nukem. Every new discovery on that Power Mac 9500 felt like unlocking a superpower. I thought the possibilities were endless.

That same childlike excitement has returned, but this time it feels different. The endless possibilities I thought I had on that Mac were not so limitless, as it turns out. But today, with the breakthroughs in deep learning and neural networks powering this recent wave of AI tooling, it actually feels like the possibilities are truly endless. We’re not just getting new tools — we’re witnessing a fundamental shift in how humans interact with computers. That’s why I’m embarking on a 100-day journey to deeply understand AI from the ground up.

Read more »

I can’t stop thinking about AI agents.

We’re witnessing the dawn of truly autonomous digital beings — software that not only follows instructions, but actually thinks, decides, and acts on its own. The implications are mind-blowing, and we’re just starting to get a real-world glimpse of what this future might look like.

When AI agents join forces

AI agents, or task-oriented autonomous agents, are already being deployed across industries — from customer support to healthcare — but their true potential lies not in individual tasks, but in collaboration. These digital workers can transform productivity when working alone, but the real revolution begins when they can negotiate and work together.

Read more »

During a recent conference, I noticed a difference in the levels of AI tool adoption between folks at startups vs. enterprises. Startups are free to explore, and are experimenting rapidly. PMs and engineers elsewhere are struggling with stringent data restrictions, lengthy approval processes, and limited tool access. Any free exploration is left to personal time. It seems like most teams are hampered out there: according to IBM, 75% of companies face barriers to adoption. [1]

This gap isn’t just about access, it’s about the pace of innovation. 

Read more »

Summer is ending and with it, our second year of camping with a little one. Turns out, traveling with a curious and energetic 18-month-old reveals some valuable insights. Here's what we've learned about how to pack smart and efficiently for family camping adventures:

What to Pack

Read more »

If you are in the business of software integrations, your API is your brand’s digital handshake with the world. How you present your API and documents directly reflects your values as a business. If you keep details clear and familiar, you'll have developers singing your praises. Overcomplicate it, and you’re basically paving the way for your competitors.

I’ve been on both sides of the equation, as a builder and customer. I’ve built and launched APIs for multi-sided marketplaces and designed large-scale operations systems automated entirely by API. I've felt the thrill of a seamless integration and the frustration of a poorly organized one. Trust me when I say this: a little added care in your API design goes a long way. It can be the factor that differentiates you from competition.

Who is this essay for? The PM or Engineering Lead who is managing / launching APIs and dealing with technical constraints, scalability, and user experience.

I get how complicated integrations can be. Your API might need multi-level schemas or intricate procedures. That complexity doesn’t have to bleed into your design and documentation. Remember why you built an API in the first place? To make your app easier to use. So let's keep things simple for your most important customer, the developer.

Read more »

For July 4th this year, I went out with some friends to Lake Washington. We rented a boat for a few hours, directly from this couple living right on the water. The rental experience was a delight: quick, convenient, and a price we felt great about. Having rented a boat online in a similar peer-to-peer way in 2022, I asked to see if was the same business (my friend had rented the boat) - it wasn't.

It felt similarly convenient to what I love about AirBnb: you drive up and get settled right away. For what seems like a great business, there haven't been any runaway successes yet. I thought, "What's up with that?"

Seems there are a few peer-to-peer boat rental companies, but nobody dominates. See: Boatsetter (2012), GetMyBoat (2013), Click&Boat (2013), Borrow A Boat (2014).

Read more »

Product testing exists to make sure that things we build, work as intended. If something about your customer’s experience is broken, you might lose them forever. Companies obviously want people to use their products reliably and avoid this — product testing makes for a logical solution.

Yet despite its importance, product testing has plenty of issues. Software teams are routinely subject to short timelines (QA teams regularly hear ”Can you test this last minute?”). Regression testing, which makes sure new code does not impact existing functionality, is often the first thing cut when resources are tight. This combination of resource constraints plus technical debt have made adoption of best practices (like continuous testing) and new AI / ML techniques, damn near impossible for most companies to implement.

When over half of software companies suck at something they all acknowledge is critical, something is amiss. Either people are lying about how much they value product testing or the model is broken.

Read more »