Teresa Torres on Claude Code for Non-Technical PMs
Product discovery expert Teresa Torres reveals how she uses Claude Code for task management, research automation, and writing—without coding experience.
Why Product Managers Should Consider Claude Code
Teresa Torres, author of the widely-read Continuous Discovery Habits, joins Claire Vo on How I AI to demonstrate something unexpected: a product manager with minimal coding experience using Claude Code as her primary productivity tool—for everything except coding.
On discovering pair programming for everything: "I pair program now with everything I do, even if it's not programming. I pair task manage and I pair write and I pair everything." Torres's insight is that the collaborative model developers use with AI coding assistants translates directly to knowledge work.
On AI seeing your work: "By moving my task management to Claude, now Claude sees my tasks and I can literally start my day and be like, 'Claude, what's on my to-do list that you can just do for me?'" The key shift isn't just using AI—it's giving AI visibility into your actual work context.
On data portability: Torres's journey started with anxiety about vendor lock-in: "As time went on, I just started to get really worried about how am I ever going to get my data out of Trello." By moving to markdown files with Claude Code, she owns her data completely while gaining AI capabilities.
On context architecture: "To do context well, it's not just that we have to document everything. We have to document everything in teeny-tiny files so that when we ask Claude to do a task, we can give Claude just the context it needs." This is the key insight—granular context files beat monolithic documentation.
On lazy prompting: Torres built extensive context libraries so she can be "lazy with prompting." Her business profile, writing style guide, and product documentation let her simply say "Claude, blog post review" and get feedback calibrated to her actual goals.
5 Workflows That Make Claude Code Practical for PMs
- Custom task management - A
/todayslash command generates daily to-do lists from markdown files, checking Trello boards and surfacing overdue items automatically - Research automation - Daily searches of arXiv and Google Scholar, with overnight AI summaries of any papers she downloads
- Context libraries - Indexed folders of business context, personal info, and writing style guides that Claude loads based on the task
- Intelligent search - Finding things across notes even when she remembers them wrong: "I have a thing called new blog post tomorrow" finds "article Wednesday"
- Writing partnership - Real-time research while drafting, section-by-section feedback, and typo fixes—all without Claude actually writing for her
What This Means for AI-Augmented Knowledge Work
Torres articulates a framework many professionals are discovering: the question isn't automation vs. doing it yourself, but "automation or augmentation" for every task. She deliberately chooses to keep writing while automating research summaries—AI handles what she doesn't enjoy while she keeps what she does.
The deeper insight is about identity and craft: "I love to write. I don't want to automate writing." As AI capabilities expand, knowledge workers are learning to be intentional about which parts of their work define them versus which parts merely need to get done.
