> ## Content Index
> Fetch the complete content index at: https://productengineers.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# How to Become a Product-Minded Engineer in the AI Era
- URL: https://productengineers.com/episodes/season2/how-to-become-a-product-minded-engineer-in-the-ai-era/
- Published: 2026-09-08T10:56:02.000Z
- Updated: 2026-09-08T10:56:02.000Z
- Description: Drew Hoskins is a senior staff software engineer at Stripe and the author of O’Reilly’s The Product-Minded Engineer. After more than 20 years in engineering at companies including Microsoft, Meta, and Stripe, he moved into product management at Temporal before returning to engineering.
- Author: Drew Hoskins
- Tags: #season-2

## Episode notes.

Drew Hoskins is a senior staff software engineer at Stripe and the author of O’Reilly’s *The Product-Minded Engineer*. After more than 20 years in engineering at companies including Microsoft, Meta, and Stripe, he moved into product management at Temporal before returning to engineering.

In this conversation, we explore how engineers can make more impactful decisions without becoming product managers. Drew explains how to uncover customer needs, use detailed scenarios to expose gaps and design better tests, build tools that sustain product thinking, and prepare for an AI-enabled future of end-to-end software generalists.

---

Listen on [YouTube](https://youtu.be/JtUfbaVdP%5FM?ref=productengineers.com), [Spotify](https://t.co/fT5CBaKQDM?ref=productengineers.com) or [Apple Podcasts](https://t.co/1T0sqvFZfJ?ref=productengineers.com)

---

### In our conversation, we discuss:

1. What product-minded engineering means and why teams need both a system view and a user-story view
2. How product judgment helps engineers choose higher-impact companies, teams, and technical investments
3. How to learn from sales calls, support conversations, GitHub issues, product data, and direct customer contact
4. Why detailed user scenarios reveal plot holes, guide prioritization, and become stronger end-to-end tests
5. How AI could create software generalists who own the complete product cycle while specialists tackle harder problems

### Where to find Drew Hoskins:

- [LinkedIn](https://www.linkedin.com/in/drewhoskins2/?ref=productengineers.com)
- [The Product-Minded Engineer newsletter](https://drewhoskins.substack.com/)

---

### Lessons Learned

- Ask why before touching the ticket. An impact statement gives you the context to resolve missing requirements, make better trade-offs, and avoid implementing the request too literally.
- Hold the system view and the story view together. Components, databases, and interfaces explain how software works, while the sequence of user actions reveals whether the complete experience works.
- Treat developers as real users. APIs and internal platforms require discovery, clear concepts, safe defaults, useful errors, and thoughtful interfaces just as customer-facing products do.
- Seek customer evidence where it already lives. Search sales recordings, support transcripts, Slack conversations, and GitHub issues when direct access to users is difficult.
- Instrument the user journey. Track activation, retention, churn, and where people abandon a flow—not only application errors—so you can distinguish acquisition problems from usability or value problems.
- Write detailed user scenarios. Describe who the user is, what motivates them, how they discover and understand the feature, what they do, how the system responds, and what happens next.
- Hunt for plot holes. When a scenario assumes that users magically find a feature, understand an unfamiliar concept, or skip an unexplained step, treat that gap as a product problem to investigate.
- Turn important scenarios into tests. Exercise connected steps together so the test proves that the real journey works across pages, links, state, cookies, and other integration boundaries.
- Prioritize impact over technical fascination. An elegant optimization creates little value when it addresses no meaningful bottleneck, while product context directs engineering effort toward outcomes users notice.
- Build tools before expecting culture to change. Searchable customer knowledge, useful metrics, and strong test infrastructure make product-minded behavior a natural part of everyday engineering work.
- Scrutinize AI-generated tests. Tests are the auditor of the product, but an agent can produce shallow assertions that merely confirm its own implementation. Review whether they prove the intended user outcome.
- Preserve collaboration as ownership expands. AI may let one engineer carry discovery, specification, implementation, and delivery, but colleagues remain essential for challenging bias, sharing context, and preventing solitary rabbit holes.