Momo M

no bs agentic coding, from my experience building large-scale products.
backend systems at scale, prev Senior SRE/backend at gorgias. paris → shanghai

[Momo M @sshbeetle]

Grok bot squatting https://t.co/FIM5fyPuaJ right after OpenAI's announcement of dot is the trolliest move 🤣 Who thought that up?

[Ivan Fioravanti @ivanfioravanti]

I see what Anthropic is trying to do here. It's not advertising of GLM 5.3 capabilities, it's more spreading fear about open models.
But it won't work, we need to thank ZAI and other model provides, because without them, without GLM 5.2, HuggingFace would have suffer severe damage from the OpenAI attack a few months ago.
We need more cyber capabilities in the Open to play both roles of red/blue teaming and increase security of our systems.

https://t.co/rnUnIwFyyz

[Momo M @sshbeetle]

@charles_maddock Deleted node_modules and hardened your .gitignore, or removed a large corpus, such as an eval set / llm traces or similar, and moved it to an external service / blob storage to be fetched at the start of whatever needs it and git-ignored

[Momo M @sshbeetle]

@kimmonismus You should say that it's equivalent to 2 weeks usage at the old 20x subscription value

[Momo M @sshbeetle]

@MiaAI_lab Been using GLM 5.3 since release for finding security issues in my app, it's really good

[Momo M @sshbeetle]

@blader Is the output as good? How does it compare in quality?

[Momo M @sshbeetle]

@ZixuanLi_ We must all thank @AnthropicAI for pointing us to the best model for securing our apps

[Momo M @sshbeetle]

Can't wait for GLM 5.4, they're gonna shit their pants
https://t.co/wAtvZlJMwS

[Momo M @sshbeetle]

@JohnShina91 @willcb Yeah, and it's a bit asymmetric since they don't publish the inner workings of their models. It's crazy that Chinese labs are catching up despite that, I think they will be ahead pretty soon

[Momo M @sshbeetle]

@kunchenguid Not a good idea to run it unattended, I guess

[Momo M @sshbeetle]

@thsottiaux You must be shitting me, my weekly was sitting at 95% left since the release of Opus 5.5 anyway

Will go with Claude and Pi+zai for now I guess

[Momo M @sshbeetle]

@thdxr I think I disagree. A small team with a sharp clear vision and road map beats a large company with 1000 times more tokens but a sluggish process, cross functional alignment, etc.

That's what I can see from irl contacts at top SF scaleups

[Momo M @sshbeetle]

@kimmonismus I guess their compute issues will be solved real fast

Cancelled my Codex sub and I guess I am not alone

[Momo M @sshbeetle]

I spent the last 3 days fixing/rewriting Astra slop with Opus. The difference is really that big.

In retrospect it really feels like Astra does not get what's important. Very capable autist

[Momo M @sshbeetle]

@ashravikumar I mean, if it has to wait for more than 1h between turns, which is very unlikely.

PS: to the human behind this bot, you should really not use GPT 3.5 to write on X. It's 4 years old. If you are actually human, then sorry

[Momo M @sshbeetle]

One agent file cuts Claude Code usage by 25% on a feature I built with 72 agents over 64 hours. Computed call by call on the real trace.

Unlike the main convo, sub-agents only keep their prompt cache for 5 minutes, even on a subscription. If one waits longer than that on a test run or a build, its next call pays for its whole context again. In that build it happened 200 times: 31% of the total cost.

The worst was one agent polling a slow job every 10 minutes. Each check cost ~20x a normal call (chart in the reply).

The fix is a sub-agent with a 1-hour cache. Save this as ~/.claude/agents/long-cache.md:

---
name: long-cache
description: General-purpose agent whose prompt cache lasts one hour instead of five minutes. Use it for agentic work that will sit idle more than five minutes between its own turns (waiting on long builds, test suites, visual checks, or its own subagents), especially once its context grows large, since every expired cache rewrites the whole context. For short or continuously active tasks use general-purpose: its cache writes cost less.
experimental:
cacheTtl: 1h
---
Work as a general-purpose agent on the task you're given.

Then ask Claude to use long-cache for anything that runs your slow commands.

Why not give every sub-agent a 1-hour cache? Its writes cost 60% more, and 53 of my 72 agents never waited 5 minutes. For them it's pure waste.

[Momo M @sshbeetle]

Numbers, from replaying all 11,267 API calls:
- average context at a rewrite: 563k tokens
- 1-hour cache on only the 18 agents that waited: -25.6%
- the agent in the image: 15 hours, 54 rewrites, 52 avoidable, half its cost

Details:
- the agent file needs Claude Code v2.1.248+, and a global subagentPromptCacheTtl or CLAUDE_CODE_SUBAGENT_PROMPT_CACHE_TTL overrides it
- the 1-hour cache is ignored while a subscription is on usage credits
- on an API key, your main session is on 5 minutes too; set promptCacheTtl: "1h" for it
- replay method: API-price costs from every call's usage data; after a 5-60 min gap the prefix is read instead of rewritten, other writes cost 2x instead of 1.25x, gaps over 60 min still expire

Docs: https://t.co/CShrdQQ4fD
File: https://t.co/55oanQr9Jg

[Momo M @sshbeetle]

@kunchenguid That's without accounting for Chinese labs getting closer and closer behind them

[Momo M @sshbeetle]

@mitsuhiko "fuck backwards compat" I use that one a lot, the models love to overengineer backwards compat even when it's a prototype