One Hour PM

I'm building a library - a work in progress which I'm calling One Hour PM - that uses generative AI agents to help product managers accelerate value creation and capture. It does this by exploring how agents and automations can improve creative and productive ways of working with product ideas and tasks.

The library has the following key features:

Simple: One input interface across all tasks
Multimodal: Inputs and outputs work with many data types
Curated: Modern product tools and templates selected with a bias for lean workflows
Contextual: Add custom data sources for more specific and specialised outcomes
Connected: Access to internal and external tools
Automatic: Agents execute creative and directed tasks semi-autonomously
Fast: Compressing task cycle times to minutes and workflow cycle times to an hour
Solo: Get more done with only yourself-in-the-loop
Safe: Validated output that enforces structure and reliability

Tips and Tricks

Stage	Tips & Tricks
Planning & Design	- Iterative Loop: • Chat: Explore options • Play: Edit in OpenAI playground (60% time) • Loop: Add data and test cases • Nest: Simplify with subtasks - Time Allocation: • Playing: 60% • Prompt Tuning: 20% • Input Massaging: 10% • Coding: 10% • Tooling: 1%
Development	- Dos: • Use all modalities • Code for input and output structure • Leverage pretraining information • Reduce search space upfront - Don'ts: • Add abstractions between yourself and the LLM • Stick to one model • Have too high an I/O ratio
Testing & Debugging	- Debugging: • Start at the prompt level or try a different model • Transform input and add more structure to output • Classify tasks and errors to identify where the failure is - Optimizing AI: • Break prompts to reduce complexity • Use separate models for structure and long-form writing • Implement state management and self-healing
Deployment & Scaling	- Future Planning: • Assume everything will be 10x to 50x cheaper and faster in the future • Build for at least 6 months ahead - AI Resources: • Be mindful of the exponential scaling of attention algorithms