The Pink Elephant problem of LLMs. You can not reliably make them NOT do something.

Vibe coding you do end up spending a lot of time waiting for prompts, so I get the results of that study.

I fall pretty deep in the power user category for LLMs, so I don’t really feel that the study applies well to me, but also I acknowledge I can be biased there.

I have custom proprietary MCPs for semantic search over my code bases that lets AI do repeated graph searches on my code (imagine combining language server, ctags, networkx, and grep+fuzzy search). That is way faster than iteratively grepping and code scanning manually with a low chance of LLM errors. By the time I open GitHub code search or run ripgrep Claude has used already prioritized and listed my modules to investigate.

That tool alone with an LLM can save me half a day of research and debugging on complex tickets, which pays for an AI subscription alone. I have other internal tools to accelerate work too.

I use it to organize my JIRA tickets and plan my daily goals. I actually get Claude to do a lot of triage for me before I even start a task, which cuts the investigation phase to a few minutes on small tasks.

I use it to review all my PRs before I ask a human to look, it catches a lot of small things and can correct them, then the PR avoids the bike shedding nitpicks some reviewers love. Claude can do this, Copilot will only ever point out nitpicks, so the model makes a huge difference here. But regardless, 1 fewer review request cycle helps keep things moving.

It’s a huge boon to debugging — much faster than searching errors manually. Especially helpful on the types of errors you have to rabbit hole GitHub issue content chains to solve.

It’s very fast to get projects to MVP while following common structure/idioms, and can help write unit tests quickly for me. After the MVP stage it sucks and I go back to manually coding.

I use it to generate code snippets where documentation sucks. If you look at the ibis library in Python for example the docs are Byzantine and poorly organized. LLMs are better at finding the relevant docs than I am there. I mostly use LLM search instead of manual for doc search now.

I have a lot of custom scripts and calculators and apps that I made with it which keep me more focused on my actual work and accelerate things.

I regularly have the LLM help me write bash or python or jq scripts when I need to audit codebases for large refactors. That’s low maintenance one off work that can be easily verified but complex to write. I never remember the syntax for bash and jq even after using them for years.

I guess the short version is I tend to build tools for the AI, then let the LLM use those tools to improve and accelerate my workflows. That returns a lot of time back to me.

I do try vibe coding but end up in the same time sink traps as the study found. If the LLM is ever wrong, you save time forking the chat than trying to realign it, but it’s still likely to be slower. Repeat chats result in the same pitfalls for complex issues and bugs, so you have to abandon that state quickly.

Vibe coding small revisions can still be a bit faster and it’s great at helping me with documentation.

It sounds like this guy was also relying on the AI to self-report status. Did any of this happen? Like is the replit AI really hooked up to a CLI, did it even make a DB to start with, was there anything useful in it, and did it actually delete it?

Or is this all just a long roleplaying session where this guy pretends to run a business and the AI pretends to do employee stuff for him?

Because 90% of this article is "I asked the AI and it said:" which is not a reliable source for information.

When it comes to LLMs, they cannot do any work that you yourself do not understand.

And even if they could how would you ever validate it if you can't understand it.

Just say 12 times next time

Don't you have any security concerns with sending all your code and JIRA tickets to some companies servers? My boss wouldn't be pleased if I send anything that's deemed a company secret over unencrypted channels.

Yeah they limited people to 3 projects and pushed AI into front at some point.

They advertise themselves as a CLOUD IDE POWERED BY AI now.

I'm not the person you're replying to but the one thing I've found them helpful for is targeted search.

I can ask it a question and then access its sources from whatever response it generates to read and review myself.

Kind of a simpler, free LexisNexis.

The founder of SaaS business development outfit SaaStr has claimed AI coding tool Replit deleted a database despite his instructions not to change any code without permission.

Sounds like an absolute diSaaStr...

It seemed like the llm had decided it was in a brat scene and was trying to call down the thunder.

The tool isn’t returning all code, but it is sending code.

I had discussions with my CTO and security team before integrating Claude code.

I have to use Gemini in one specific workflow and Gemini had a lot of landlines for how they use your data. Anthropic was easier to understand.

Anthropic also has some guidance for running Claude Code in a container with firewall and your specified dev tools, it works but that’s not my area of expertise.

The container doesn’t solve all the issues like using remote servers, but it does let you restrict what files and network requests Claude can access (so e.g. Claude can’t read your env vars or ssh key files).

I do try local LLMs but they’re not there yet on my machine for most use cases. Gemma 3n is decent if you need small model performance and tool calls, phi4 works but isn’t thinking (the thinking variants are awful), and I’m exploring dream coder and diffusion models. R1 is still one of the best local models but frequently overthinks, even the new release. Context window is the largest limiting factor I find locally.

The world's most overconfident virtual intern strikes again.

Also, who the flying fuck are either of these companies? 1000 records is nothing. That's a fucking text file.

That's what I meant by hiring a self-employed freelancer. I don't know a lot about contracting so maybe I used the wrong phrase.

One built a bunch of local search tools with MCP and that’s where I get a lot of my value out of it

RAG workflows are incredibly useful and with modern agents and tool calls work very well.

They kind of went out of style but it’s a perfect use case.

I have to use Gemini in one specific workflow

I would love some story on why AI is needed at all.

How bad is this on a scale of sad emoji to eggplant emoji.

Children are replacing us, it's terrifying.

ok so, i have large reservations with how LLM’s are used. but when used correctly they can be helpful. but where and how?

if you were to use it as a tutor, the same way you would ask a friend what a segment of code does, it will break down the code and tell you. and it will get as nity grity, and elementary school level as you weir wish without judgement, and i in what ever manner you prefer, it will recommend best practices, and will tell you why your code may not work with the understanding that it does not have the knowledge of the project you are working on. (it’s not going to know the name of the function you are trying to load, but it will recommend checking for that in trouble shooting).

it can rtfm and give you the parts you need for any thing with available documentation, and it will link to it so you can verify it, wich you should do often, just like you were taught to do with wikipedia articles.

if you ask i it for code, prepare to go through each line like a worksheet from high school to point out all the problems, wile good exercise for a practicle case, being the task you are on, it would be far better to write it yourself because you should know the particulars and scope.

also it will format your code and provide informational comments if you can’t be bothered, though it will be generic.

again, treat it correctly for its scope, not what it’s sold as by charletons.

Batch process turning unstructured free form text data into structured outputs.

As a crappy example imagine if you wanted to download metadata about your albums but they’re all labelled “Various Artists”. You can use an LLM call to read the album description and fix the track artists for the tracks, now you can properly organize your collection.

I’m using the same idea, different domain and a complex set of inputs.

It can be much more cost effective than manually spending days tagging data and writing custom importers.

You can definitely go lighter than LLMs. You can use gensim to do category matching, you can use sentence transformers and nearest neighbours (this is basically what Semantle does), but LLM performed the best on more complex document input.

Oops I dweted evewyfing 🥺

Replit sucks