Ryan NisbetOperations, Delphi · Operations at Delphi

What Jason Lemkin Learned from His Delphi

5 Jun 2025 · 1 min

Watch on YouTube

1:12

In this segment of the YouTube video, Jason Lemkin explains how he built an AI assistant—referred to as Delphi—by ingesting a massive personal corpus of roughly 10,000 pieces of his own content (tweets, LinkedIn posts, SaaStr videos) together with thousands of interview transcripts from prominent SaaS founders such as Dharmesh Shah, Brian Halligan, Yamini Rangan, and Aaron Levie. He describes the ingestion pipeline, which relies on simple web‑scraping techniques: hooking up RSS feeds, pulling data via APIs, and providing URLs for the scraper. Lemkin notes that the AI quickly outperforms his own memory because it can recall every detail he has ever published. After experimenting with a few approaches, he settled on using the Delphi platform to orchestrate the data collection and model serving. The overall narrative emphasizes the practical steps of aggregating personal and third‑party content, the technical simplicity of the ingestion process, and the surprising performance boost of a personal AI over the creator’s own recall.

Chapters

for like seeing the future a little bit is I have about 10,000 pieces of content I've written over the years, a lot of it which is not out of date, and I update it personally. So about 10,000 pieces of content. And then we have several thousand interviews like you guys do, but with so many great folks, early ones with Dharmesh, middle ones from Dharmesh, Dharmesh when, when HubSpot had just IPO'd, Dharmesh the other day, Yamini last week, Brian Halligan over the years. And we have all of it. And our AI took everything I've written every tweet I've ever done, every LinkedIn post, every SaaStr video from every speaker for 12 years. Aaron Levy, the week he went IPO for the first SaaStr Annual to last week when we were talking about AI, all of it ingested in this AI. And so what I instantly saw was as soon as I would talk to this AI, it was much better than me. It was much better than me because I forget stuff. How did you get that information in there? You just hook up RSS, right? Right? The feed for a blog. It can scrape the web, which is pretty easy. Soon we'll be able to do other things like MCB, but it's just scraping APIs and RSS. It's really just scraping. You just give a URL. But did you use like Delphi or you, you built your own tool or what'd you do? I tried a couple of things and then I used Delphi.