1. Did you have MCP tooling setup so it can get lsp feedback? This helps a lot with code quality as it’ll see warnings/hints/suggestions from the lsp

2. Unit tests. Unit tests. Unit tests. Unit tests.

I cannot stress enough how much less stupid LLMs get when they jave proper solid Unit tests to run themselves and compare expected vs actual outcomes.

Instead of reasoning out “it should do this” they can just run the damn test and find out.

They’ll iterate on it til it actually works and then you can look at it and confirm if its good or not.

I use Sonnet 4.5 / 4.6 extensively and, yes, its prone to getting the answer almost right but a wrong in the end.

But the unit tests catch this, and it corrects.

Example: I am working on my own fame engine with monogame and its about 95% vibe coded.

This transform math is almost 100% vibe coded: https://github.com/SteffenBlake/Atomic.Net/blob/main/MonoGame/Atomic.Net.MonoGame/Transform/TransformRegistry.cs

The reason its solid is because of this: https://github.com/SteffenBlake/Atomic.Net/blob/main/MonoGame/Atomic.Net.MonoGame.Tests/Transform/Integrations/TransformRegistryIntegrationTests.cs

Also vibe coded and then sanity checked by me by hand to confirm the math checks out for the tests.

And yes, it caught multiple bugs, but the agent automatically could respond to that, fix the bug, rerun the tests, and iterate til everything was solid.

Test Driven Development is huge for making agents self police their own code.

The difference, when the tool is used correctly, is so massive that only someone deeply uninformed or naive would contend it.

I got about 4 entire days worth of work completed in about 5 hours yesterday at my job, thats just objective fact.

Tasks that used to take weeks now take days, and tasks that used to take days now take hours. Theres no “feeling” about this, Ive been a software developer for approaching 17 years now professionally. I know how long it takes to produce an entire gambit of integration tests for a given feature. I spend almost all of my time now reviewing mountains of code (which is fairly good quality, the machines produce fairly accurate results), and then a small amount of time refining it.

People deeply do not at all understand how dramatically the results have changed over the past 2 years, and their biases are based on how things were 2 years ago.

Sure, 2 years ago the quality was way worse, the security was bad, the enforcement almost non existent, and peoples overall skill with how to use the tools was just beginning to grow. You cant exactly be good at using a tool that only just came out.

But its been two years of very rapid improvement. Its good now. Anyone who has been using these tools and actually monitoring progression can speak to this.

Things heavily shifted about 5 months ago when competition started to really fire up between different providers, and I wont say its even close to great yet, but its definitely good, it works, its fast, and it’s pretty damn good at what I need it to do.

You know programmers who use llms believe they’re much more productive because they keep getting that dopamine hit, but when you actually measure it, they’re slower by about 20%.

Everyone keeps citing this preliminary study and ignores:

1. Its old now
2. Its sample size was incredibly tiny
3. Its sample group were developers not using proper tooling or trained on how to use the tools

Its the equivalent of taking 12 seasoned carpenters with very little experience on industrial painting, handing them industrial grade paint guns that are misconfigured and uncalibrated, and then asking them to paint some of their work and watching them struggle… and then going “wow look at that industrial grade paint guns are so bad”

Anyone with any sense should look at that and go “thats a bogus study”

But people with intense anti-ai bias cling to that shoddy ass study with such religious fervor. Its cringe.

Every professional developer with actual training and actual proper tooling can confirm that they are indeed tremendously more productive.

Lovely anthropic mcp. Make sure you give anthropic lots of money and use their tools

Its becoming clear you have no clue wtf you are talking about.

Model Context Protocol is a protocol, like http or json or etc.

Its just a format for data, that is open sourced and anyone can use. Models are trained to be able to invoke MCP tools to perform actions, and anyone can just make their own MCP tools, its incredibly simple and easy. I have a pretty powerful one I personally maintain myself.

Anthropic doesnt make any money off me, in fact, I dont use any of their shit, except maybe whatever licensing fees microsoft pays to them to use Claude Sonnet, but microsoft copilot is my preferred service I use overall.

I bet you your contract with them says they’re not liable for shit their llm does to your files

Setting aside the fact that I dont even use anthropic’s tools, my copilot LLMs dont have access to my files either. Full stop.

The only context in which they do have access to files is inside of the aforementioned docker based sandbox I run them inside of, which is an ephemeral immutable system that they can do whatever the fuck they want inside of because even if they manage to delete /var/lib or whatever, I click 1 button to reboot and reset it back to working state.

The working workspace directory they have access to has readonly git access, so they can pull and do work, but they literally dont even have the ability to push. All they can do is pull in the stuff to work on and work on it

After they finish, I review what changes they made and only I, the human, have the ability to accept what they have done, or deny it, and then actually push it myself.

This is all basic shit using tools that have existed for a long time, some of which are core principles of linux and have existed for decades

Doing this isnt that hard, its just that a lot of people are:

1. Stupid
2. Lazy
3. Scared of linux

The concept of “make a docker image that runs an “agent” user in a very low privilege env with write access only to its home directory” isnt even that hard.

It took me all of 2 days to get it setup personally, from scratch.

But now my sandbox literally doesnt even expose the ability to do damage to the llm, it doesnt even have access to those commands

Let me make this abundantly clear if you cant wrap your head around it:

LLM Agents, that I run, dont even have the executable commands exposed to them to invoke that can cause any damage, they literally dont even have the ability to do it, full stop

And it wasnt even that hard to do

You’ll be the 4753rd guy with the oops my llm trashed my setup and disobeyed my explicit rules for keeping it in check

Read what I wrote.

Its not a matter of “rules” it “obeys”

Its a matter of literally not it even having access to do such things.

This is what Im talking about. People are complaining about issues that were solved a long time ago.

People are running into issues that were solved long ago because they are too lazy to use the solutions to those issues.

We now live in a world with plenty of PPE in construction and people are out here raw dogging tools without any modern protection and being ShockedPikachuFace when it fails.

The approach of “Im gonna tell the LLM not to do stuff in a markdown file” is tech from like 2 years ago.

People still do that. Stupid people who deserve to have it blow up in their face.

Use proper tools. Use MCP. Use a sandbox environment. Use whitelist opt in tooling.

Agents shouldn’t even have the ability to do damaging actions in the first place.

The only people who have these issues, are people who are using the tools wrong or poorly.

Using these models in a modern tooling context is perfectly reasonable, going beyond just guard rails and instead outright only giving them explicit access to approved operations in a proper sandbox.

Unfortunately that takes effort and know-how, skill, and understanding how these tools work.

And unfortunately a lot of people are lazy and stupid, and take the “easy” way out and then (deservedly) get burned for it.

But I would say, yes, there are safe ways yo grant an llm “access” to data in a way where it does not even have the ability to muck it up.

My typical approach is keeping it sandbox’d inside a docker environment, where even if it goes off the rails and deletes something important, the worst it can do is cause its docker instance to crash.

And then setting up via MCP tooling that commands and actions it can prefer are explicit opt in whitelist. It can only run commands I give it access to.

Example: I grant my LLMs access to git commit and status, but not rebase or checkout.

Thus it can only commit stuff forward, but it cant even change branches, rebase, nor push either.

This isnt hard imo, but too many people just yolo it and raw dawg an LLM on their machine like a fuckin idiot.

These people are playing with fire imo.

Very true, though theres a certain threshold you can get past where the context, at least, is usable in size where the machine can at least hold enough data at once for common tasks.

One of the pieces of tech we are really missing atm is an automation of being able to filter info.

Specifically, for the LLM to be able to “release” info as it goes asap as unimportant and forget it, or at least it gets stored into some form of long term storage it can use a tool to look up.

But for a given convo the LLM can do a lot of reasoning but all that reasoning takes up context.

Itd be nice if after it reasons, it then can discard a bunch of the data from that and only keep what matters.

This eould tremendously lower context pressure and allow the LLM to last way longer memory wise

I think tooling needs to approach how we manage LLM context in a very different way to make further advancement.

LLMs have to be trained to have different types of output, that control if they’ll actually remember it or not.

You: just the cheap ones

I never said that. I just said that the cheap ones are especially shitty.

People on this site really lack reading comprehension it seems.

LLMs are not good at answering fact based questions, fundamentally. Unless its an incredibly well known answer that has never changed (like a math or physics question), they dont magically “know” things.

However, they’re way better at summarizing and reasoning.

Give them access to playwright web search capability via MCP tooling to go research info, find the answer(s), and then produce output based on the results, and now you can get something useful.

“Whats the best way to do (task)” << prone to failure, functional of how esoteric it is.

“Research for me the top 3 best ways to do (task), report on your results and include your sources you found” << actually useful output, assuming you have something like playwright installed for it.

What are you talking about.

No? I never said that.

I just explained /why/ it happened, I literally nowhere in my post said, or implied, someone should pay for more expensive models. What are you smoking?

You just have to be aware they have very short memory when using a cheap model and assume anything you wrote 1 minute ago has already left its memory, which is why they produce pretty dumb output if you try and depend on that… so… dont depend on that.

Uh… no its just the free models being free, theyre lower cost intentionally to provide free options for people who dont wanna pay subscription fees.

(context is (V)RAM)

Eh sort of, its more operating costs, the larger the context size the more expensive the model is to run, literally in terms of power consumption.

Keep in mind we are on the scale of fractions of cents here, but multiply that by millions of users and it adds up fast.

But the end result is that the agent will fuck stuff up, and will even quickly /forget/ it fucked that up if you dont catch it asap

A lot of them have a context window that can be wiped out within like, 2 minutes of steady busywork…

They dont lol

Pretty much always this is just the fact cheaper, especially free, chatbots, have very limited context windows.

Which means the initial restrictions you set like “dont do this, dont touch that” etc get dropped, the LLM no longer has them loaded. But it does have in the past history the very clear and urgent directives of it trying to do this task, its important, so it’ll do whatever it autocompletes its gotta do to accomplish the task. And then… fucks something up.

When you react to their fuck up, it *reloads the context back in

So now the LLM has in its history just this:

1. It doing a thing against the rules
2. The user yelling at it
3. The users now getting loaded after that on top

So now the LLM is going to autocomplete its generated text on top being very apologetic and going on about how it’ll never happen again.

Thats all there is to it.

Itd be weird to be exempt

Its really sounding like the FCC banned all routers and companies can only comply by running their own they install software on manually XD

TIL! I assume theres gotta be at least some US manufacturing of routers… I hope… otherwise the US just banned using the internet in effect lol

Isn’t TP Link a US company?

Basically yep.

No they mean routers.

And running your own router with its own OS is allowed.

Buying a prefab router from Shenzhen chock full of foreign malware is not.

Only locally produced malware is now allowed in prefab routers!

Folks definitely should be picking up learning how to setup their own router and wifi hubs, but they wont do it.

Its more and more seeming less like “Fuvk AI” and more and more like “I am bad at using AI and Im gonna blame my skill issues on everyone else”

This entire post sits in pure “git gud” territory, wait-

git: unrecognized command "good"

Dang

…wat?

LLMs are completely capable of invoking npm install, the fuck you smoking lol.

Theyre perfectly able to read docs and use existing libraries fine.

Arguably I find better, because when I bump into a bug in a lib, I can fork it, fix the bug, get my patch up, and use the patch in my project in like 15min flat, no longer having to even worry about “ugh is it worth the effort to fix it”

Yes, its so easy now. I just do it.

Furthermore agents can look up documentation in mere seconds and read it (better than Id say a lot of junior devs tend to approach problems ngl)

If the agent is equipped with adequate tools and instructions, its extremely productive under strict guidance.

The same way.

The result the LLM produces is a link to the relevant information directly I can click and go to it.

Example would be a giant collection of files, think like, 10gb+ of many pdfs, or more. I want any relevant sections on a topic, it quickly can aggregate on it and give me links I click to open up straight to relevant sections in specific files, and then read rhem.

This stuff comes up a lot in my industry (software dev) as we often inherit huge data pools of legacy documentation on massive codebases.

When I am tasked with fixing 1 small specific piece of the system, it could take me hours to find the specific stuff Im looking for on the (often poorly maintained) docs.

But also vector db setup to map to that data, and an LLM wired up to it, can search it in milliseconds and pull up relevant sections asap, and I can click and dig deeper from there as much as I need.

This sort of “fuzzy searching” vectorization of tokens is what an LLM does very well. Its part of how it produces its output, but you can reverse the process to create search indexes (effectively reversing the data through the LLM to turn the data into deterministic vectors)

Amd its important to note, the fault tolerance you perceive here doesnt apply. Thos specific type of searching with vector DBs will always produce the same results from the same input, everytime. Its deterministic.