> In the past year, agent harnesses crossed the “can it be done” rubicon.

Brother, I'm still in "Can you get it right?"-mode. What am I doing wrong? (Rhetorical, but advice welcomed).

dosisking 24 minutes ago | parent | next [-]

There are two 'camps' with respect to AI.

One camp already knows that Neural Nets don't work and are a dead end.

The other camp hasn't yet figured out that Neural Nets don't work, but are convinced that they do (or eventually will), because they think everything always improves over time in a linear fashion.

▲

jaggederest 41 minutes ago | parent | prev | next [-]

I'd be happy to screenshare with you if you like, we can work on something trivial or open source. Half an hour should be more than enough to see whether you're doing anything obviously self-sabotaging.

▲

al_borland 3 hours ago | parent | prev | next [-]

I’ve found some success is small projects, with limited scope, in a greenfield.

I’m terrified to attempt agentic anything in the repo my job actually cares about. I triggered it once by accident, when the agent was first rolled out and enabled by default… it broke everything. Now I just use ask mode, and even that is wrong half the time, and once it goes wrong it just keeps getting worse.

I saw a post from Dave Plumber who vibe coded up a new cross platform task manager. He said his spec document for the AI was 107 pages long. So maybe what I’m doing wrong is not giving the AI a literal novel of spec.

▲

applfanboysbgon 2 hours ago | parent [-]

> He said his spec document for the AI was 107 pages long.

This sounds like programming but with extra steps that make it take longer with less reliability.

	▲	0x696C6961 an hour ago \| parent [-]
		Ikr, at that point the code itself is a better way of encoding the information.

▲

mw888 an hour ago | parent | prev | next [-]

You're appealing to ambiguity. All you've said is you have failed—how is anyone supposed to know what went wrong?

▲

simonw 2 hours ago | parent | prev | next [-]

Tell it to use red/green TDD and start things off with an already configured test suite, maybe with a single test that asserts 1+1==2.

Make sure it know how to run the tests before it starts writing any additional code.

Then set it a clear goal.

▲

slopinthebag 2 hours ago | parent [-]

Basically all the examples of LLM's building impressive things have been because they have human written tests to base the implementation on. If you have an LLM write the tests the results are far less impressive or valuable.

▲

bharatsuthar an hour ago | parent [-]

Yes and LLMs are known to cheat on tests written by them.

	▲	slopinthebag an hour ago \| parent [-]
		It's not always cheating either. They aren't intelligent, so they don't actually understand the purpose of the tests or can build them to define the actual semantics of the problem space. It's literally just next-token prediction based on the codebase and prompt. Cheating implies that they have agency, and ironically agents don't.

▲

MattGaiser 3 hours ago | parent | prev [-]

What is “it” specifically and what languages are you using?