I would have loved to see more of the harness engineering shared as code. Instead we are left with only the outcome.

I guess that makes sense since the harness is the product in the case of Cursor.

dakolli 21 hours ago | parent | next [-]

I call this junk "meta-agentic engineering", it reminds me of people who have the coolest nvim configs, spend hundreds of hours customizing it but ultimately get less work done than the guy with minimal workflows, if any at all.

I look on twitter and it's just people building tools for agents to use agents, some weird customization loop going on in the LLM space right now. Ultimately these are trends pushed on us from model providers because they 10x token consumption. Its literally just BS trends to increase revenue at these companies, most of it is largely useless.

▲

svachalek 20 hours ago | parent | next [-]

Some of this stuff is ludicrous. I finally tried /loop last week and discovered every loop iteration passes the entire context history to the model. So pretty quickly you're running a full 1M context window, without cache, likely just to check if something is ready or needs to be done. It's miserably terrible engineering unless your entire and only goal is to burn tokens.

▲

Wowfunhappy 20 hours ago | parent [-]

Wait, why isn't context cached with /loop?

▲

pzh 19 hours ago | parent | next [-]

Because it's typically cached for 5min (1hr is a setting you have to explicitly opt into), and very few people run loops at a cadence of < 5 mins.

▲

EliasWatson 2 hours ago | parent | next [-]

https://code.claude.com/docs/en/prompt-caching#on-a-claude-s...

"On a Claude subscription, Claude Code requests the one-hour TTL automatically. Usage is included in your plan rather than billed per token, so the longer TTL costs you nothing extra and only affects how long your cache stays warm. If you’ve gone over your plan’s usage limit and Claude Code is drawing on usage credits, you are billed for that usage, so Claude Code automatically drops the TTL to five minutes."

▲

Wowfunhappy 19 hours ago | parent | prev [-]

Oh! I thought Anthropic cached for one hour (by default), am I wrong about that? Or is this an OpenAI thing?

	▲	poly2it 18 hours ago \| parent [-]
		Anthropic changed their cache duration for some reason a while back.

▲

ihsw 20 hours ago | parent | prev [-]

[dead]

▲

brap 6 hours ago | parent | prev | next [-]

I think many of us engineers (including myself) are suffering from AI-induced procrastination. Fixing that pesky bug in production isn't fun. Building a harness/workflow that detects and fixes pesky bugs in production on its own is way more fun (especially when your employer pays for tokens), and it feels just as important, if not more! So what if it didn't actually fix that pesky bug in production on its own...

▲

ubercore 4 hours ago | parent | prev [-]

I never liked the term "ricing" with window managers, but this is definitely veering into LLM ricing territory sometimes. And people are getting rich off of it.

▲

blmarket 19 hours ago | parent | prev [-]

I don't think there is a lot test harness, given there is a golden implementation (sqlite3), with clear interface (SQL, and sqlite3 db files). Once you slop crappy implementation, then you can create one group of agents finding gap where slop doing wrong vs golden, and another group of agents filling the gap meeting the conformance.

I assume they did similar, as long as test case(or spec) can be parallelized. Having parallel execution / testing environment is crucial if we need 1000 commits/second throughput.

I would not surprised they have 1+ millions of specs.

I would bet they won't release generated specs - actually this will be a valuable asset like https://sqlite.org/th3.html