Pi Durable(earendil.com)
474 points by paulsmith | 65 comments
Earendil announcement of Pi 1.0 alongside Pi Durable, an experimental package for long-running, durable, malleable AI agents that can run anywhere.
474 points by paulsmith | 65 comments
Earendil announcement of Pi 1.0 alongside Pi Durable, an experimental package for long-running, durable, malleable AI agents that can run anywhere.
708 points by apitman | 203 comments
An Earendil engineering post arguing that inference API sessions are becoming non-portable, so users can no longer own or move their conversation transcripts between models.
1618 points by sergiotapia | 553 comments
Earendil announces the 1.0 release of Pi, a hardened, minimal, extensible AI agent harness used by hundreds of thousands of people weekly.
487 points by luispa | 257 comments
Earendil post arguing that Pi, a minimal AI coding harness, gains performance and cost advantages by avoiding the complexity of larger agent tools.
238 points by doppp | 222 comments
193 points by tosh | 81 comments
372 points by nadis | 351 comments
Quandri engineering blog post by backend engineer Chloe Kim titled "MCP is dead," arguing about the Model Context Protocol for AI agents.
708 points by apitman | 203 comments
An Earendil engineering post arguing that inference API sessions are becoming non-portable, so users can no longer own or move their conversation transcripts between models.
413 points by gmays | 335 comments
A blog post arguing that the Model Context Protocol (MCP) is preferable to agent skills for extending LLM tools.
341 points by virtualwhys | 109 comments
Andy Pavlo blog post from April 2023 criticizing PostgreSQL's multi-version concurrency control (MVCC) implementation and its drawbacks.
733 points by marcjschmidt | 206 comments
Ask HN post by the Deepkit maintainer describing how a VC-funded company got his EU trademark revoked for insufficient proof of genuine use, seeking advice.
668 points by skilldeliver | 165 comments
Medium article by Elena Cross arguing that the Model Context Protocol (MCP) for AI agents has significant security weaknesses.