This is completely excessive, but we threw our AI at solving fantasy sports analysis and added a bunch of native integrations and python libraries. It's pretty decent, and fun to just chat with the data.
Big props to Joseph Wilson (@pseudo-r) who maintains the unofficial ESPN API. https://github.com/pseudo-r/Public-ESPN-API. Reliably the best service for interacting with fantasy sports data and AI.
This was an interesting project to work on, mainly because the quality and popularity of API docs for various services really does impact the experience of untrained LLMs using them out of the box. It was fun to experience this outside of our regular B2B context, and my big takeaway from this project was just how much people are underestimating the importance of this as we experience the rise of the agentic web over the next 3 years. (Agents are users too!)
We included some easter egg cricket features, but we haven't found a free service the AI likes enough to bundle that in officially yet, so that'll come at a later date. (We're Aussies, so of course the fantasy Cricket analysis will improve over time : )
mceoin•5mo ago
Big props to Joseph Wilson (@pseudo-r) who maintains the unofficial ESPN API. https://github.com/pseudo-r/Public-ESPN-API. Reliably the best service for interacting with fantasy sports data and AI.
This was an interesting project to work on, mainly because the quality and popularity of API docs for various services really does impact the experience of untrained LLMs using them out of the box. It was fun to experience this outside of our regular B2B context, and my big takeaway from this project was just how much people are underestimating the importance of this as we experience the rise of the agentic web over the next 3 years. (Agents are users too!)
We included some easter egg cricket features, but we haven't found a free service the AI likes enough to bundle that in officially yet, so that'll come at a later date. (We're Aussies, so of course the fantasy Cricket analysis will improve over time : )