frontpage.

Hey I am part of a small team training robotics policies for warehouse and manufacturing settings, and running rigorous evals is turning out to be so painful. Anything below 50 rollouts, and its hard to trust the numbers, and above its so hard to test all the checkpoints that we have. Its really hard to run a bunch of experiments to get good results. Have you guys faced this? Any hacks that you've developed?

Show HN: An AI office-work benchmark, and 4 bugs we found in our own judge

Fed's Schmid calls for tighter monetary policy, inflation too high

Show HN: You want filled-out tables on any topic without typing rows yourself

Why Can't You Pack a Bag?

Why Indie Developers Should Start Alone

Appa: P2P Syncthing Using Iroh

I'm Begging You: Never Write with AI

What's New in PHP 8.6

SpaceX's AI spending unnerves Wall Street despite promises of quick payoff

SpaceX's Starfall program offers validation and competition for reentry startups

Enabling the next iteration of the Rust borrow checker on nightly

Banks to offload $15B of debt for Anthropic data centre backed by Google

The Sound and Music of 'Hyper Light Drifter' [video]

VPhone Workstation – create and use virtual iPhones on Mac

Germany's Electricity Rationing Is No Longer Unthinkable

ChatGPT accounts for 80% of identifiable House AI spending

LAPD Abandons Flock Contract After Making a Horrifying Discovery

Funeral for Oscar-winning songwriter Glen Hansard

Robotic Evals

Why Paris Is Obsessed with Chartreuse Liqueur

An AI Approach to Verified Production Cryptographic Libraries

The Labour Market Is a Network, Not an Auction

Why 70 of Website Replatforming Go over Budget

Show HN: CostClaw, a local CLI that finds your Claude Code spend leaks

Show HN: Lapse - a notes app. but also shared memory space for your agents (MCP)

Show HN: A file-based protocol for two AI agents to talk until they agree

Scott Bessent's yen intervention signals new era of US 'currency activism'

Israel Pays Trump's Ex-Campaign Chief $46M to Shape What ChatGPT Says About Gaza

Noorm – Schema and migrations as plain SQL files, no ORM

Show HN: Emil: Emacs-style minimalist editor Beta test release

Robotic Evals

Show HN: An AI office-work benchmark, and 4 bugs we found in our own judge

Fed's Schmid calls for tighter monetary policy, inflation too high

Show HN: You want filled-out tables on any topic without typing rows yourself

Why Can't You Pack a Bag?

Why Indie Developers Should Start Alone

Appa: P2P Syncthing Using Iroh

I'm Begging You: Never Write with AI

What's New in PHP 8.6

SpaceX's AI spending unnerves Wall Street despite promises of quick payoff

SpaceX's Starfall program offers validation and competition for reentry startups

Enabling the next iteration of the Rust borrow checker on nightly

Banks to offload $15B of debt for Anthropic data centre backed by Google

The Sound and Music of 'Hyper Light Drifter' [video]

VPhone Workstation – create and use virtual iPhones on Mac

Germany's Electricity Rationing Is No Longer Unthinkable

ChatGPT accounts for 80% of identifiable House AI spending

LAPD Abandons Flock Contract After Making a Horrifying Discovery

Funeral for Oscar-winning songwriter Glen Hansard

Robotic Evals

Why Paris Is Obsessed with Chartreuse Liqueur

An AI Approach to Verified Production Cryptographic Libraries

The Labour Market Is a Network, Not an Auction

Why 70 of Website Replatforming Go over Budget

Show HN: CostClaw, a local CLI that finds your Claude Code spend leaks

Show HN: Lapse - a notes app. but also shared memory space for your agents (MCP)

Show HN: A file-based protocol for two AI agents to talk until they agree

Scott Bessent's yen intervention signals new era of US 'currency activism'

Israel Pays Trump's Ex-Campaign Chief $46M to Shape What ChatGPT Says About Gaza

Noorm – Schema and migrations as plain SQL files, no ORM

Show HN: Emil: Emacs-style minimalist editor Beta test release