I don't trust anyone who sees the output of current generation of LLMs and thinks "I want that to be agentic!". It can be immensely useful, but it's only useful when it manages to make your own brain work more, not less.
I’m finding agentic coding to be a fascinating tool. The output is a mess but it takes so little input to make something quite functional. I had an app that I wrote with a python GUI framework I didn’t quite like. ChatGPT rewrote it to use GTK and it is so much faster now. Later Claude added a browser mode where the app can be run via GTK or a browser tab. I have never written a GTK app in my life past some hello world text box.
The output is very problematic. It breaks itself all the time, makes the same mistakes multiple times, I have to retread my steps. I’m going to have it write tests so it can better tell what it’s breaking.
But being able to say “take this GTK app and add a web server and browser based mode” and it just kinda does it with minimal manual debugging is something remarkable. I don’t fully understand it, it is a new capability. I do robotics and I wish we had this for PCB design and mechanical CAD, but those will take much longer to solve. Still, I am eager to point Claude at my hand written python robotics stack from my last major project [1] and have it clean up and document what was a years long chaotic prototyping process with results I was reasonably happy with.
The current systems have flaws but if you look at where LLMs were five years ago and you see the potential value in fixing the flaws with agentic coding, it is easy to imagine that those flaws will be addressed. There will be higher level flaws and those will eventually be addressed, etc. Maybe not, but I’m quite curious to see where this goes, and what it means for engineering as a human being at these times.
It is fascinating and it absolutely excels at writing barely-working, problematic code, that yet somehow appears to run. This helps me a lot, as having a shitty code to fix makes my mind much more engaged than when I'm writing stuff from scratch, but making the model do more stuff autonomously rather than having me consciously review it at each step is only making it less useful, not more.
As I see it, all of civilization is built on top of this "laziness" principle of "I'm tired of having to deal with X, how can I sort X so that I don't need to think about it anymore, and can instead focus on Y". I'm general, people want their brain to work on other stuff, not what's currently in front of them.
It's exactly what agentic tools make harder to do. LLM-generated code usually looks great at the first glance - your opinion on how bad it is is a function of effort spent reviewing, analyzing and questioning it.
The shitty code it comes up with helps me a lot, because fixing broken stuff and unraveling the first principles of why it's broken is how I learn best. It helps me think and stay focused. When learning new areas, the goal is to grasp the subject matter enough to find out what's wrong with the generated code - and it's still a pretty safe bet that it will be wrong, one way or another. Whenever I attempt to outsource the actual thinking (because of feeling lazy or even just to check the abilities of the model), the results are pretty bad and absolutely nowhere near the quality level of anything I'd want to sign my name under.
Of course, some people don't mind and end up wasting other people's time with their generated patches. It's not that hard to find them around. I have higher standards than replying "dunno, that's what AI wrote" when a reviewer asks why is something done this particular way. Agentic tools bring down the walls which could let you stop for a moment and notice the sloppiness of that output even further. They just let the model do more of the things it's not very good at, and encourage it to flood the code with yet another workaround for an issue that would disappear completely had you spent two minutes pondering about the root cause (which you won't do, because you don't have a mental model of what the code does, because you let the agent do it for you).
ChatGPT agent has now been made available in the EU and all other supported countries and territories for Pro, Plus, Team, Enterprise, and Edu plans, as was to be expected [0].
> I have it available already for one or two weeks.
I think I've had it available with the separate website ("research preview"?) for months, but yeah, last few weeks it's been directly in ChatGPT.com, and I'm within the EU.
Yea, I was looking forward to test new gpt5, but giving them my ID + scanning my face is not something I am willing to do to be allowed to give them my money.
An agent as far as I've seen people use it is a script that will add some stuff to your prompt and monitor the LLM's output for a specific pattern and execute code when it encounters that.
For instance, you could have an "agent" that can read/edit files on your computer by adding something like "to read a file, issue the `read_file $path`" to your prompt, and whenever a line of LLM output that starts with `read_file` is finished, the script running on your computer will read that file, paste it into the prompt, and let the LLM continue its autocomplete-on-steroids.
If you write enough tools and a complicated enough prompt, you end up with an LLM that can do stuff. By default, smart tools usually require user confirmation before actually doing stuff, but if you run the LLM in full agent mode, you trust the LLM not to do anything it shouldn't. curl2bash with LLMs, basically.
An LLM with significant training and access to file access, HTTP(S) API access, and access to some OS APIs can do a lot of work for you if you prompt it right. My experience with Claude/Copilot/etc. is that 75% of the time, the LLM will fail to do what it should be doing without manually repairing its mistakes, but in the other 25% of the time it does look rather sci-fi-ish.
With some tools you can tell your computer "take this directory, examine the EXIF data of each image, map the coordinates to the country and nearest town the picture was taken in, then make directories for each town and move the pictures to their corresponding locations". The LLM will type out shell commands (`ls /some/directory`), interpret the results as part of the prompt response that your computer sends back, and repeat that until its task has been completed. If you prepare a specific prompt and set of tools for the purpose of managing files, you could call that a "file management agent".
Generally, this works best for things you can do by hand in a couple of minutes or maybe an hour if it's a big set of images, but something the computer can now probably take care of you for you. That said, you're basically spending enough CO2 to drive to the store and back, so until we get more energy efficient data centers I'm not too fond of using these tools for banal interactions like that.
An agent (in the way the term is commonly currently used around LLMs) is a combination of LLM, external tools, and management framework such that the total system can make (and make use of the results of) one or multiple multiple tool calls at the LLM direction without intervening user interaction to serve user needs. (Usually, in practice, this takes place in between the request and response in what is otherwise a typical chatbot-style interaction, though there are other possibilities.)
Think of an agent as a standalone script or service. They have a single function take inputs and create outputs.
You can chain agents together into a string to accomplish larger tasks.
Think of everything involved in booking travel. You have set a budget, pick dates, chose a destinations, etc…. Each step can be defined as an agent and then you chain them together into a tool that handles the entire task for you.
Agent originally meant an ai that made decisions to optimize some utility function. This was seen as a problem: we don’t know how to pick a good utility function or even how to point an ai at a specific utility function, so any agent that was smarter than us was as likely as not to turn us all into paperclips, or carve smiles into our faces, or some other grim outcome.
With LLMs, this went through two phases of shittifaction: first, there was a window where the safety people were hopeful about LLMs because the weren’t agents, so everyone and their mother declared that they would create an agent out if an LLM explicitly because they heard it was dangerous.
This pleased the VCs.
Second, they failed to satisfy the original definition, so they changed the definition of agent to the thing that they made and declared victory. This pleased the VCs
This isn't strictly speaking true. An agent is merely something that acts (on its environment). A simple reflex agent (eg. simple robot vacuum with only reflexive collision detection) are also agents, though they don't strictly speaking attempt to maximize a utility function.
In other words, VC-backed tech companies decided to weaken the definition of 'Torment Nexus' after they failed to create the Torment Nexus inspired by the classic sci-fi novel 'Don't Create the Torment Nexus'.
Readit News
The output is very problematic. It breaks itself all the time, makes the same mistakes multiple times, I have to retread my steps. I’m going to have it write tests so it can better tell what it’s breaking.
But being able to say “take this GTK app and add a web server and browser based mode” and it just kinda does it with minimal manual debugging is something remarkable. I don’t fully understand it, it is a new capability. I do robotics and I wish we had this for PCB design and mechanical CAD, but those will take much longer to solve. Still, I am eager to point Claude at my hand written python robotics stack from my last major project [1] and have it clean up and document what was a years long chaotic prototyping process with results I was reasonably happy with.
The current systems have flaws but if you look at where LLMs were five years ago and you see the potential value in fixing the flaws with agentic coding, it is easy to imagine that those flaws will be addressed. There will be higher level flaws and those will eventually be addressed, etc. Maybe not, but I’m quite curious to see where this goes, and what it means for engineering as a human being at these times.
[1] https://github.com/sequoia-hope/acorn-precision-farming-rove...
The shitty code it comes up with helps me a lot, because fixing broken stuff and unraveling the first principles of why it's broken is how I learn best. It helps me think and stay focused. When learning new areas, the goal is to grasp the subject matter enough to find out what's wrong with the generated code - and it's still a pretty safe bet that it will be wrong, one way or another. Whenever I attempt to outsource the actual thinking (because of feeling lazy or even just to check the abilities of the model), the results are pretty bad and absolutely nowhere near the quality level of anything I'd want to sign my name under.
Of course, some people don't mind and end up wasting other people's time with their generated patches. It's not that hard to find them around. I have higher standards than replying "dunno, that's what AI wrote" when a reviewer asks why is something done this particular way. Agentic tools bring down the walls which could let you stop for a moment and notice the sloppiness of that output even further. They just let the model do more of the things it's not very good at, and encourage it to flood the code with yet another workaround for an issue that would disappear completely had you spent two minutes pondering about the root cause (which you won't do, because you don't have a mental model of what the code does, because you let the agent do it for you).
[0] https://news.ycombinator.com/item?id=44596320
Tried it once and it really sucked.
I think I've had it available with the separate website ("research preview"?) for months, but yeah, last few weeks it's been directly in ChatGPT.com, and I'm within the EU.
What is ChatGPT Agent?
https://news.ycombinator.com/item?id=44837367
Just give us the API and stop trying OpenAI.
Is for example Google’s crawl bot an agent?
Is there a prominent successful agent that I could test myself?
So many questions…
For instance, you could have an "agent" that can read/edit files on your computer by adding something like "to read a file, issue the `read_file $path`" to your prompt, and whenever a line of LLM output that starts with `read_file` is finished, the script running on your computer will read that file, paste it into the prompt, and let the LLM continue its autocomplete-on-steroids.
If you write enough tools and a complicated enough prompt, you end up with an LLM that can do stuff. By default, smart tools usually require user confirmation before actually doing stuff, but if you run the LLM in full agent mode, you trust the LLM not to do anything it shouldn't. curl2bash with LLMs, basically.
An LLM with significant training and access to file access, HTTP(S) API access, and access to some OS APIs can do a lot of work for you if you prompt it right. My experience with Claude/Copilot/etc. is that 75% of the time, the LLM will fail to do what it should be doing without manually repairing its mistakes, but in the other 25% of the time it does look rather sci-fi-ish.
With some tools you can tell your computer "take this directory, examine the EXIF data of each image, map the coordinates to the country and nearest town the picture was taken in, then make directories for each town and move the pictures to their corresponding locations". The LLM will type out shell commands (`ls /some/directory`), interpret the results as part of the prompt response that your computer sends back, and repeat that until its task has been completed. If you prepare a specific prompt and set of tools for the purpose of managing files, you could call that a "file management agent".
Generally, this works best for things you can do by hand in a couple of minutes or maybe an hour if it's a big set of images, but something the computer can now probably take care of you for you. That said, you're basically spending enough CO2 to drive to the store and back, so until we get more energy efficient data centers I'm not too fond of using these tools for banal interactions like that.
You can chain agents together into a string to accomplish larger tasks.
Think of everything involved in booking travel. You have set a budget, pick dates, chose a destinations, etc…. Each step can be defined as an agent and then you chain them together into a tool that handles the entire task for you.
Deleted Comment
Deleted Comment
With LLMs, this went through two phases of shittifaction: first, there was a window where the safety people were hopeful about LLMs because the weren’t agents, so everyone and their mother declared that they would create an agent out if an LLM explicitly because they heard it was dangerous.
This pleased the VCs.
Second, they failed to satisfy the original definition, so they changed the definition of agent to the thing that they made and declared victory. This pleased the VCs
Ref: Artificial Intelligence - A Modern Approach.
Can’t say much about the usage as I haven’t tried it yet.