He built Claude Code a memory so he could trust its advice
A developer built a persistent memory system for Claude Code using markdown files stored locally on their computer, enabling the AI agent to maintain context across multiple sessions and provide better-grounded advice based on accumulated knowledge about the user.
Summary
The speaker explains their motivation for creating a memory system for Claude Code, driven by a desire to ensure that AI advice is grounded in accurate, personalized information about them. Rather than relying solely on the chat interface, they switched to using a coding agent approach. The system works by having Claude Code read and write markdown files that serve as persistent memory, stored locally on the speaker's computer. This allows the memory to persist between separate chat sessions—when the speaker returns and starts a new Claude Code session, the AI can reference the same files and maintain continuity of context. The speaker identifies two major benefits from this implementation: first, it significantly increased their confidence in the AI's advice because the system is grounded in verified personal information; second, it improved the AI's ability to create actionable outputs like checklists, which helps keep the speaker motivated and moving forward on tasks. The system essentially creates a bridge between separate AI sessions, preventing information loss and enabling more sophisticated, personalized interactions over time.
Key Insights
- The speaker built a markdown file read/write system in Claude Code specifically to ground AI advice in verified personal information, rather than relying on the AI's general knowledge
- The persistent memory system stores markdown files locally on the user's computer, enabling Claude Code to maintain context and reference the same information across multiple separate chat sessions
- Implementing this persistent memory system significantly increased the speaker's confidence in the AI's advice because it ensures the AI is working from accurate, personal ground truth
- The system improved the AI's ability to generate actionable outputs like checklists that keep the user motivated and moving forward on tasks
- The speaker chose to implement this system by using a coding agent approach rather than the standard chat interface, demonstrating a preference for programmatic control over memory and context management
Topics
Transcript
[0:00] I wanted to build a system where I was confident that AI was grounded in my truth and knowing what it knows about me so that I could be confident in what advice it was giving me back. So what I did instead of just going with the chat interface, switching over to a coding agent, so in cloud code, I built this system that would read and write markdown files so they would have the memory of what we were talking about and it would be living on my computer so that if I come back later and start a new chat session, a cloud code session, [0:31] it can reference those same files. This has been a…
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
Fraudsters wrote PRDs on how to scam my AI product
An AI product founder shares critical lessons about preventing fraud and abuse of AI services, emphasizing that fraudsters actively create detailed plans to exploit AI platforms. Key defenses include monitoring costs instead of request counts, verifying clients thoroughly, and implementing strict policies with no exceptions.
I use Stripe Radar to catch card testing before it costs me
The speaker explains how to use Stripe Radar to prevent card testing fraud by leveraging built-in fraud rules, fingerprint detection, and Spot ID verification. These tools help identify suspicious payment patterns and validate that requests originate from legitimate app users, while allowing tuning to minimize false positives for good users.
Token fraudsters relayed their AI product through ChatPRD
The founder of Chat Purity discusses token fraud, where bad actors authenticate as their service and relay requests from their own AI products through Chat Purity to appear legitimate. Despite existing technical controls, fraudsters employed reverse proxies and relay systems to obscure their activity, creating inefficient and suspicious infrastructure.
How I fight token fraud as an AI founder
Claire, founder of ChatPRD, discusses the widespread problem of token fraud—where bad actors abuse free trials, promo codes, and other mechanisms to steal API inference at the expense of AI product companies. She shares six types of abuse she's experienced, specific fraud examples, and details both technical and operational countermeasures she's implemented to protect her bootstrapped business.
He wants AI that keeps students doing the cognitive effort in math
The speaker emphasizes that AI in mathematics education should support student agency and cognitive effort rather than providing direct answers. Effective AI tools should interpret student ideas, encourage deeper thinking, and engage students in productive struggle while keeping them as active decision-makers in their learning process.