How I turned Claude into a self-improving PM assistant | Daniel Blum (PM, Melio) · How I AI summary

Key takeaways

Building a systematic, self-improving AI infrastructure can dramatically increase productivity and accuracy in daily work.
Fundamental rules for powerful AI systems include the ability to self-rewrite and deep integration with various tools and data sources.
Automations like weekly prep and daily briefs can serve as central hubs for managing complex workflows, continuously learning, and updating for better performance.
Scaling a personal AI system to a team requires accessible UX, shared resources, and a strong onboarding process to multiply its benefits.
Despite advancements, key challenges remain, such as creating fully autonomous AI that can operate independently outside of local devices, which holds the potential for even greater productivity gains.

Summary

Revolutionizing PM Workflows with AI

Daniel Blum shares his journey of building a highly systematic AI-driven productivity ecosystem that automates and manages core aspects of his daily work as a product manager. Central to this is his custom Notion board and various automation tools like Cowork and Claude, which handle everything from weekly planning to daily briefs. This approach has allowed him to accomplish in a day what used to take a week, emphasizing the importance of a well-structured system rather than just the choice of tools.

Core Principles and Building Blocks of an AI System

The episode highlights two fundamental rules that make AI systems powerful: 1) The ability to rewrite and improve their own core logic continuously, and 2) Deep integration with a broad ecosystem of tools. Daniel discusses how these principles guide his setup, enabling his AI tools to learn, adapt, and handle complex workflows by contextualizing internal terminology and automating routine tasks.

Deep Dive into Daniel’s AI Workflow and Automation

Daniel elaborates on his weekly and daily automation routines, such as the weekly prep that populates his Notion board by aggregating insights from Slack, email, and meetings. His morning brief then provides summaries and flags unknown terms for clarification, helping him maintain context and focus. The system is designed to learn from interactions, flag frictions, and suggest new skills, creating a self-improving loop that enhances his productivity over time.

Scaling and Extending the System Across the Organization

Daniel describes how he scaled his personal AI system to his team through a plugin called Workstation, which streamlines onboarding and ensures consistent setup for all employees in about 15 minutes. The approach emphasizes accessible UX and shared ownership, making advanced AI workflows usable for non-technical team members while allowing continuous customization and improvement.

Challenges and Future of AI in Productivity

The podcast addresses remaining gaps, such as the need for a more autonomous, always-on AI that can operate independently in the cloud outside of local device constraints. Daniel reflects on the ongoing effort to push AI toward a more autonomous future, enabling even greater efficiency. He also discusses how this new level of automation frees him to focus on high-value tasks like customer research and product development.

Show notes

Daniel Blum is a product manager at Melio, a B2B payments company, and one of the most systematic thinkers I’ve had on the show when it comes to personal AI infrastructure. He’s spent the past year building a Claude- and Cowork-based productivity system that manages his Notion board, processes his Slack and email, and runs self-improvement loops every week without needing to be prompted. Beyond his own workflow, Daniel built and scaled a “Workstation” onboarding plugin that gets any Melio employee up and running with a personalized Claude setup in about 15 minutes.

What you’ll learn:

Why Daniel says the two rules that make any AI system powerful aren’t about the tool you pick
How his weekly prep automation fills an entire Notion board from scratch every Sunday, without his touching it
The morning brief feature that teaches Claude new internal terms on its own, so company jargon never slows it down
Why he describes Notion as “read-only” now, and what that says about how PM workflows are changing
The self-improvement loop that watches Daniel’s edits, spots recurring friction, and suggests new skills to build
How he uses a skill called “Improve” to filter the endless flood of AI tips without drowning in them
What he built to scale his personal system to every PM at Melio, and the UX lesson he learned the hard way
The capability gap that’s still keeping him from running 100% of his work through Claude

—

Brought to you by:

Optimizely—Your AI agent orchestration platform for marketing and digital teams

Jira AI SDLC—Get your tokens’ worth with Jira

—

In this episode, we cover:

(00:00) Daniel’s background and the PM overhead problem he needed to solve

(03:30) His AI stack at Melio

(05:00) The two rules that make any AI system genuinely powerful

(06:00) The Notion board Cowork built for him (and manages on his behalf)

(07:30) How he contextualizes Claude with voice memos, links, and recurring updates

(09:00) His weekly prep automation