We’re introducing HALO 😇 Hierarchal Agent Loop Optimizer HALO is an RLM-based agent optimization technique capable of recursively self-improving agents by analyzing their execution traces and suggesting changes. This work is inspired by the Mismanaged Genius Hypothesis proposed by @a1zhang and @lateinteraction earlier this month. tldr; we improved performance on AppWorld (Sonnet 4.6) from 73.7 --> 89.5 ( 15.8) by giving HALO-RLM access to harness trace data and asking it to identify issues. The feedback from HALO surfaced failures in the harness such as hallucinated tool calls, redundant arguments in tools, refusal loops, and semantic correctness issues. Each issue mapped cleanly to a direct prompt update. We then fed these finding into Cursor (Opus 4.6), and asked the coding agent to update the underlying harness. We repeated this trace -> HALO-RLM analysis -> code update loop until the score plateaued. Today we’re open-sourcing the core HALO-RLM framework, evals, and data for further review.

This is what I wanted agent observability to feel like. Traces that end in a code change.

Get paid to wait The Claude Code spinner might be the most watched line on Earth. So I turned it into an ad marketplace. Advertisers bid on it. You keep 50% of the money. Install the extension → get cash from ads. Introducing Kickbacks

I sat down with @samhogan He raised $11M to build private AI infrastructure. Then he asked for another $11M. He says companies should stop sending data to OpenAI. Naturally, I asked about SOC 2. He had one. This was upsetting. I prefer when founders are easier to prosecute.

Startup idea: Retatrutide blow darts. We have to start forcibly lowering the obesity rate.

Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore. You should be designing loops that prompt your agents.