How (and why) to build a personal API
The conversation explores the concept of personal APIs that store individual preferences like favorite restaurants and coffee orders, enabling others to make thoughtful gestures without asking. The discussion evolves to envision AI agents autonomously accessing these APIs to perform personalized tasks like making reservations.
Summary
The speaker introduces the concept of a personal API as a solution to a common social problem: wanting to do something nice for someone (like making a restaurant reservation) but lacking specific knowledge about their preferences. Rather than asking directly and spoiling the surprise, a personal API would store this information in an accessible format.
The conversation then shifts to a more forward-looking perspective on how this concept would actually be used in practice. While the initial idea assumes humans would manually query the API, the respondent argues that the more realistic future involves AI agents performing this work autonomously. They paint a scenario where an agent, given a simple verbal instruction after a podcast episode, would independently access the personal API, retrieve relevant information like travel dates and dining preferences, and execute actions like restaurant reservations without human intervention.
The discussion touches on how this concept integrates with broader trends in agentic commerce, suggesting that personal APIs could become a key component in a larger ecosystem where AI agents handle increasingly complex, personalized transactions on behalf of users. The exchange indicates growing interest in exploring what kinds of APIs and data structures could enable this future of agent-driven personalized services.
Key Insights
- The original motivation for personal APIs stems from wanting to perform thoughtful gestures for others without directly asking about their preferences, preserving the element of surprise
- The respondent challenges the assumption that humans would be the primary consumers of personal APIs, arguing that AI agents will actually be the ones accessing and utilizing this data
- The practical vision for personal APIs involves agents receiving high-level natural language instructions and autonomously executing multi-step tasks like API access, preference retrieval, and reservation booking
- Personal APIs are positioned as part of a larger ecosystem combining agentic AI with commercial transaction capabilities
- The concept bridges personal data infrastructure with emerging agent-based commerce systems, creating new opportunities for automated, personalized interactions
Topics
Transcript
[0:00] I had this idea about personal APIs. And that stemmed from this problem that you have where you want to do something nice, take them to their favorite restaurant, and make a reservation, but you don't really remember what their coffee order is or what their favorite restaurant is. And you want to do something nice, but you don't want to bug them about it. So, I was like, "Well, what if there was a way where you could just hit an API and you get all of this information about them?" >> Do you know what I love about this idea of the API? You're very adorable in that you think that humans are going to hit this…
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
Hacking an old pager with AI
A developer describes setting up an old pager as a Twitter notification device using modern cloud services. The system routes Twitter API events through Cloudflare Workers, Resend email service, and Gmail to finally deliver notifications to the pager's email address.
Opus 5: too timid?
The speaker criticizes Opus 5 for being excessively cautious and indecisive, constantly asking for human permission or approval rather than taking action. Even when faced with simple tasks like fixing a one-line merge conflict, Opus 5 deferred to the original branch owner's preferences, requiring the speaker to repeatedly push it to make autonomous decisions.
GPT-5.6's video editing via Codex is genuinely one of my favorite new workflows
A product manager describes using GPT-5.6 with Codex to automate video editing for social media clips. By simply dragging a file and providing natural language instructions, the AI generated five polished, fast-paced hype videos from a long conference talk recording, dramatically reducing the time-intensive manual clipping process.
Theoretically Intelligent vs. Practically Effective: Why GPT-5.6 Sol Beats Fable for Product Work
An executive contrasts two AI models (Fable and Soul/GPT-5.6 Sol), arguing that Soul is superior for product work because it prioritizes practical effectiveness over theoretical intelligence. The speaker values the ability to ship products to customers and understand end-user goals over theoretical sophistication.
Build a harness when the same workflow needs the same setup and the same outcomes, every time
Building a harness for repetitive workflows allows you to be more prescriptive about job execution, resulting in greater efficiency, consistency, and better outcomes. Rather than explaining requirements to an AI agent each time, a harness lets you use a simpler interface like pasting a link while the agent already understands the intended task.