I think that letting an LLM run unsupervised on a task is a good way to waste time and tokens. You need to catch them before they stray too far off-path. I stopped using subagents in Claude because I wasn't able to see what they were doing and intervene.
Indirectly asking an LLM to prompt another LLM to work on a long, multi-step task doesn't seem like a good idea to me.
I think community efforts should go toward making LLMs more deterministic with the help of good old-fashioned software tooling instead of role-playing and writing prayers to the LLM god.
Codex is like an external consultant. You give it specs and it quietly putters away and only stops when the feature is done.
Claude is built more like a pair programmer, it displays changes live, "talks" about what it's doing and what's working et.
It's really, REALLY hard to abort codex mid-run to correct it. With Claude it's a lot easier when you see it doing something stupid or getting of the rails. Just hit ESC and tell it where it went wrong (like use task build, don't build it manually or use markdownlint, don't spend 5 minutes editing the markdown line by line).
When the task is bigger than I trust the agent to work on it on its own, or for me to review the results, I ask it to create a plan with steps. Then create a md file for each step. I review the steps, and ask the agent to implement the first one. Review that one, fix it, then ask it to update the next steps, and then implement the next one. And so on, until finished.
Have you tried Scoped context packages? Basically for each task, I create a .md file that includes relevant file paths, the purpose of the task, key dependencies, a clear plan of action, and a test strategy. It’s like a mini local design doc. I found that it helps ground implementation and stabilizes the output of the agents.
Separately, you have to consider that "wasting tokens spinning" might be acceptable if you're able to run hundreds of thousands of these things in parallel. If even a small subset of them translate to value, then you're far net ahead vs with a strictly manual/human process.
I also use AI to do discrete, well-defined tasks so I can keep an eye on things before they go astray.
But I thought there are lots of agentic systems that loop back and ask for approval every few steps, or after every agent does its piece. Is that not the case?
I've tried it. It works better than raw Claude. We're working on benchmarks now. But... it's a moving target as amplifier (an experimental project) is evolving rapidly.
I’ve seen people discuss these types of approaches on X. To me it looks like the concepts here are already tried and popular - they’re just packaging it up so that people who aren’t as deep in that world can get the same benefits. But I’m not an expert.
Claude Code is great, this is just a set of tweaks, not really "research". For anyone into vibe coding, there are dozens of interesting video tutorials on customizing Claude Code and running practical jobs, not limited to coding.
I think most of us are irritated by the constant A/B Testing and underwhelming releases. Lets just have the bubble pop so we can solve real problems instead of this.
1. It affects the fundamental ego of these engineers that a computer can do what they thought only they could do and what they thought made them better than the rest of the population. They might not realize this of course.
2. AI and all these AI systems are intelligence multipliers, with a zero around IQ 100. Zero multiplied by zero is zero, and negative multiplier just leads to garbage. So the people who say "I used AI and its garbage" should really think hard about what it says about them. I thought I was crazy to think of this hypothesis but someone else also mentioned the exact statement and I didnt think I was just being especially mean anymore.
Nothing to do with ego, but you may want to check your own projections, you know how when you speak of the others, you mainly speak of yourself (Jung or Freud, not sure). No need to be bitter about not having the grind and focus to become an engineer yourself, it is after all much harder than say, earning an MBA and you should be OK with whatever you turned out to be. Not to mention that the tools themselves, were in fact built by engineers and not by the "rest of the population", like yourself. Now having said that, I am early adopter myself, was happy to pay the premium costs for my entire company, if the tool was any kind of amplifier. But the crap just does not work. Recently the quality is degrading so much that we simply reduced it to using it for simple consultation - and we only do it because unfortunately the search has been ruined. Otherwise most of the folks both internally and externally that I know using these tools would be happy to just go back to google search and SO. Unfortunately that's not an option. Also see if your second argument makes any sense at all. Maybe it comes out of a lacking math background? Firstly, you don't need two zeroes to get a zero out at the end of the multiplication. And secondly, if an average engineer is a zero, what are folks like you then? But again, it maybe just your own projections...
This is like a person who thinks that making a photocopy of an Einstein paper makes him Einstein. You know, Einstein wasn't that special after all and the photocopier affects his fundamental ego.
I've actually written my own a homebrew framework like this which is a.) cli-coder agnostic and b.) leans heavily on git worktrees [0].
The secret weapon to this approach is asking for 2-4 solutions to your prompt running in parallel. This helps avoid the most time consuming aspect of ai-coding: reviewing a large commit, and ultimately finding the approach to the ai took is hopeless or requires major revision.
By generating multiple solutions, you can cutdown investing fully into the first solution and use clever ways to select from all the 2-4 candidate solutions and usually apply a small tweak at the end. Anyone else doing something like this?
There is a related idea called "alloying" where the 2-4 candidate solutions are pursued in parallel with different models, yielding better results vs any single model. Very interesting ideas.
I've been doing something similiar: aider+gpt-5, claude-code+sonnet, gemini-cli+2.5-pro. I want to coder-cli next.
A main problem with this approach is summarizing the different approaches before drilling down into reviewing the best approach.
Looking at a `git diff --stat` across all the model outputs can give you a good measure of if there was an existing common pattern for your requested implementation. If only one of the models adds code to a module that the others do not, it's usually a good jumping off point to exploring the differing assumptions each of the agents built towards.
This reminds me of an an approach in mcmc where you run mutiple chains at different temperatures and then share the results between them (replica exchange MCMC sampling) the goal being not to get stuck in one “solution”
> "I have more ideas than time to try them out" — The problem we're solving
I see a possible paradox here.
For exploration, my goal is _to learn_. Trying out multiple things is not wasting time, it's an intensive learning experience. It's not about finding what works fast, but understanding why the thing that works best works best. I want to go through it. Maybe that's just me though, and most people just want to get it done quickly.
Yeah, this seems like the opposite of invention. You can throw paint at a canvas but it won’t make you Pollock. And will you feel a sense of accomplishment?
Hey all! I'm one of a handful of developers on this project. Great to see it's getting some interest!
For context, we are right in the middle of building this thing... multiple rebuilds daily since we are using it to build itself. The value isn't in the code itself, yet, but in the approaches (UNIX philosophy, meta-cognitive recipes, etc.)
We are really excited about how productive these approaches are even in this early stage. We are able to have amplifier go off make significant progress unattended for sometimes hours at a time. This, of course, raises a lot of questions on how software will be built in the near future... questions which we are leaning into.
Most of our team's projects, unless they have some unresolved IP or are using internal-only systems, are built in the open. This is a research project at this stage. We recognize this approach it too expensive and too hacky for most independent developers (we're spending thousands of dollars daily on tokens). But once the patterns are identified, we expect we'll all find ways to make them more accessible.
The whole point of this is to experiment and learn fast.
I do a lot of work with claude code and codex cli but frankly as soon as I see all the LLM-tells in the readme, and then all the commit messages written by claude, I immediately don't want to read the readme or try the project until someone else recommends it to me.
This is gaining stars and forks but I don't know if that's just because it's under the github.com/microsoft, and I don't really know how much that means.
When I blind-commit claude code commit messages they are sometimes totally wrong. Not even hallucinations necessarily - by the time I'm committing the context may be large and confusing, or some context lost.
I'd rather have the three word message than detailed but wrong messages.
I think I agree with you anyway on average. Most of the time a claude-authored commit message is better than a garbage message.
But it's still a red flag that the project may be filled with holes and not really ready for other people. It's just so easy to vibe your way to a project that works for you but is buggy and missing tons of features for anyone who strays from your use case.
Starting in Claude bypass mode does not give me confidence:
WARNING: Claude Code running in Bypass Permissions mode │
│ │
│ In Bypass Permissions mode, Claude Code will not ask for your approval before running potentially dangerous commands. │
│ This mode should only be used in a sandboxed container/VM that has restricted internet access and can easily be restored if damaged.
This project is a research demonstrator. It is in early development and may change significantly. Using permissive AI tools in your repository requires careful attention to security considerations and careful human supervision, and even then things can still go wrong. Use it with caution, and at your own risk.
A lot of the ideas in this aren't bad, but in general it's hacky. Context export? Just use industry standard observability! This is so bad it makes me cringe. Parallel worktrees? These are prone to putting your repo in bad states when you run a lot of agents, and you have to deal with security, just put your agent in a container and have it clone the repo. Everything this project does it's doing the wrong way.
I have a repo that shows you how to do this stuff the correct way that's very easy to adapt, along with a detailed explanation, just do yourself a favor, skip the amateur hour re-implementations and instrument/silo your agents properly: https://sibylline.dev/articles/2025-10-04-hacking-claude-cod...
I'll always be skeptical about using AI to amplify AI. I think humans are needed to amplify AI since humans are so far documented to be significantly more creative and proactive in pushing the frontier than AI. I know, it's maybe a radical concept to digest.
> I'll always be skeptical about using AI to amplify AI.
This project was in part written by Claude, so for better or worse I think we're at least 3 levels deep here (AI-written code which directs an AI to direct other AIs to write code).
I think I'm more optimistic about this than brute-forcing model training with ever larger datasets, myself. Here's why.
Most models I've benchmarked, even the expensive proprietary models, tend to lose coherence when the context grows beyond a certain size. The thing is, they typically do not need the entire context to perform whatever step of the process is currently going on.
And there appears to be a lot of experimentation going on along the line of having subagents in charge of curating the long term view of the context to feed more focused work items to other subagents, and I find that genuinely intriguing.
My hope is that this approach will eventually become refined enough that we'll get dependable capability out of cheap open weight models. That might come in darn handy, depending on the blast radius of the bubble burst.
Based on clear, operational definitions, AI is definitely more creative than humans. E.g., can easily produce higher scores on a Torrance test of divergent thinking. Humans may still be more innovative (defined as creativity adopted into larger systems), though that may be changing.
More creative? I've just seen my premium subscription "AI" struggling to find a trivial issue of a missing import in a very small / toy project. Maybe these tools are getting all sorts of scores on all sorts of benchmarks, I dont doubt it, but why are there no significant real-world results after more than 3 years of hype? It reminds of that situation when the geniuses at Google offered the job to the guy who created Homebrew and then rejected him after he supposedly did not do well on one of those algorithmic tasks (inverting a binary tree? - not sure if I remember correctly). There are also all sorts of people scoring super high on various IQ tests, but what counts, with humans as with the supposed AI is the real world results. Benchmarks without results do not mean anything.
This is absurd to the point of being comical. Do you really believe that?
If an “objective” test purports to show that AI is more creative than humans then I’m sorry but the test is deeply flawed. I don’t even need to look at the methodology to confidently state that.
> Never lose context again. Amplifier automatically exports your entire conversation before compaction, preserving all the details that would otherwise be lost. When Claude Code compacts your conversation to stay within token limits, you can instantly restore the full history.
If this is restoring the entire context (and looking at the source code, it seems like it is just reloading the entire context) how does this not result in an infinite compaction loop?
I think the idea would be that you could re-compact with a different focus. When you compact, you can give Claude instructions on what is important to retain and what can be discarded. If you later discover that actually you wanted something you discarded during a previous compaction, this could allow you to recover it.
Also, it can be useful to compact before it is strictly necessary to compact (before you are at max context length). So there could be a case where you decide you need to "undo" one of these types of early compactions for some reason.
Hi all, I'm the primary author/lead on the "research exploration" that is Amplifier at Microsoft. It's still SUPER early and we're running fast and applying learnings from the past couple of years in new ways to explore some new value we're finding early evidence of. I apologize that the repo is in a very rough condition, we're running very fast and most of what is in there now has been very helpful but will very soon be completely replaced with our next major iteration of it as we continue to run ahead. I did want to take a pause today and put together a blog post to capture a little more context for those of you here who are following along:
For those who find it useful in this very early stage, to find some value for yourself in either using it or learning from it, happy to be on the journey together. For those who don't like it or don't understand why or what we're doing, I apologize again, it's definitely not for everyone at this stage, if ever, so no offense taken.
Project looks interesting, but no demos. As much I want to try it because of all cool concepts mentioned, but I am not sure I want to invest my time if I don't see any demos
multiply it by 20 other similar projects and assume 20% have security issues, your environment will be messed up before you even understand if you need it or not. Not even talking about time you lost
>"Amplifier is a complete development environment that takes AI coding assistants and supercharges them with discovered patterns, specialized expertise, and powerful automation — turning a helpful assistant into a force multiplier that can deliver complex solutions with minimal hand-holding."
Again this "supercharging" nonsense? Maybe in Satiyas confabulated AI-powered universe, but not in the real world I am afraid...
Yeah, I'm not even that opposed to using AI for documentation if it helps, but everything from Microsoft recently has been full-on slop. It's almost like they're trying to make sure you can't miss it's AI generated.
"Eat your own dog slop" isn't bad practice, though.
Some people in the organization will experience the limitations and some will learn — although there are bound to be people elsewhere in the organization who have a vested interest in not learning anything and pushing the product regardless.
Sorry, is this Hacker News? This kind of project is exactly what I'd expect hackers to create. Not using AI in boring limited practical ways where it's known to somehow work, but supercharging AI with AI with AI... etc, and seeing what happens!
Well, stop asking silly questions. How will the execs get their bonuses, if it turns out we fucked up the web search and invested an equivalent of a moonbase in ... well, I hate to use the phrase, but statistical parrot ?
That's essentially what a CI environment does. "Multiple tabs" and "swarms". This part should feel familiar to any developer. Having multiple things running in the background to help you is not a new concept and we've been doing it for decades.
Whether these new helpers that explore ideas on their own are helpful or not, and for which cases, is another discussion.
I see you're being down voted, Reddit style. But you're on the mark about the hate tone of comments. If you don't like Amplifier, don't use it. No need to spew hate.
How can it be gatekeeping when they are literally making it easier to use? The analogy is probably closer to a Linux distro. You can put everything together yourself but if someone gives you a pre integrated environment with best practices it makes it easier to get started
Our “product” is a tool we developed internally and found it so useful that decided to open source it.
With full potential I refer to getting the best possible results. For example, being able to work on tasks in parallel without Claude instances interfering with each other vs , well, no doing so.
There are many many people who want better AI coding tools, myself included. It might or might not fail, but there is a clear and strong opportunity here, that it would be foolish of any large tech company to not pursue.
I would say it’s more the result of anti competitive bundling of cloud things into existing enterprise contracts rather than the wave. Microsoft is far worse than it ever was in the 90s but there’s no semblance of antitrust action in America.
I think that letting an LLM run unsupervised on a task is a good way to waste time and tokens. You need to catch them before they stray too far off-path. I stopped using subagents in Claude because I wasn't able to see what they were doing and intervene. Indirectly asking an LLM to prompt another LLM to work on a long, multi-step task doesn't seem like a good idea to me. I think community efforts should go toward making LLMs more deterministic with the help of good old-fashioned software tooling instead of role-playing and writing prayers to the LLM god.
There are two opposite ways to do this.
Codex is like an external consultant. You give it specs and it quietly putters away and only stops when the feature is done.
Claude is built more like a pair programmer, it displays changes live, "talks" about what it's doing and what's working et.
It's really, REALLY hard to abort codex mid-run to correct it. With Claude it's a lot easier when you see it doing something stupid or getting of the rails. Just hit ESC and tell it where it went wrong (like use task build, don't build it manually or use markdownlint, don't spend 5 minutes editing the markdown line by line).
When the task is bigger than I trust the agent to work on it on its own, or for me to review the results, I ask it to create a plan with steps. Then create a md file for each step. I review the steps, and ask the agent to implement the first one. Review that one, fix it, then ask it to update the next steps, and then implement the next one. And so on, until finished.
Have you tried Scoped context packages? Basically for each task, I create a .md file that includes relevant file paths, the purpose of the task, key dependencies, a clear plan of action, and a test strategy. It’s like a mini local design doc. I found that it helps ground implementation and stabilizes the output of the agents.
4 replies →
Separately, you have to consider that "wasting tokens spinning" might be acceptable if you're able to run hundreds of thousands of these things in parallel. If even a small subset of them translate to value, then you're far net ahead vs with a strictly manual/human process.
3 replies →
I do the same thing with my engineers but I keep the tasks in Jira and I label them "stories".
But in all seriousness +1 can recommend this method.
This is built into Cursor now with plan mode https://cursor.com/docs/agent/planning
2 replies →
this plus a reset in between steps usually helps focus context in my experience
Yeah in my experience, LLMs are great but they still need babysitting lest they add 20k lines of code that could have been 2k.
I also use AI to do discrete, well-defined tasks so I can keep an eye on things before they go astray.
But I thought there are lots of agentic systems that loop back and ask for approval every few steps, or after every agent does its piece. Is that not the case?
Please comment under this thread if you have actually tried this and can compare it to another tool like Cursor, Codex, raw Claude, etc.
I’m super not interested in hearing what people have to say from a distance without actually using it.
I've tried it. It works better than raw Claude. We're working on benchmarks now. But... it's a moving target as amplifier (an experimental project) is evolving rapidly.
A lot of snark in these comments. Has anyone actually tried it yet?
The repo is full of big AI words without any metrics/benchmark.
People are correct to question it.
If anything, Microsoft needs to show something meaningful to make people believe it's worth trying it out.
I’m not blaming them. I’m asking if anyone has tried it.
I’ve seen people discuss these types of approaches on X. To me it looks like the concepts here are already tried and popular - they’re just packaging it up so that people who aren’t as deep in that world can get the same benefits. But I’m not an expert.
Exactly. I don’t understand the cynicism in the comments and they literally are just trying to make the technology more accessible
5 replies →
Claude Code is great, this is just a set of tweaks, not really "research". For anyone into vibe coding, there are dozens of interesting video tutorials on customizing Claude Code and running practical jobs, not limited to coding.
I think most of us are irritated by the constant A/B Testing and underwhelming releases. Lets just have the bubble pop so we can solve real problems instead of this.
Hehe suddenly many people will have the real problem of paying bills unfortunately
5 replies →
I have two hypotheses:
1. It affects the fundamental ego of these engineers that a computer can do what they thought only they could do and what they thought made them better than the rest of the population. They might not realize this of course.
2. AI and all these AI systems are intelligence multipliers, with a zero around IQ 100. Zero multiplied by zero is zero, and negative multiplier just leads to garbage. So the people who say "I used AI and its garbage" should really think hard about what it says about them. I thought I was crazy to think of this hypothesis but someone else also mentioned the exact statement and I didnt think I was just being especially mean anymore.
Nothing to do with ego, but you may want to check your own projections, you know how when you speak of the others, you mainly speak of yourself (Jung or Freud, not sure). No need to be bitter about not having the grind and focus to become an engineer yourself, it is after all much harder than say, earning an MBA and you should be OK with whatever you turned out to be. Not to mention that the tools themselves, were in fact built by engineers and not by the "rest of the population", like yourself. Now having said that, I am early adopter myself, was happy to pay the premium costs for my entire company, if the tool was any kind of amplifier. But the crap just does not work. Recently the quality is degrading so much that we simply reduced it to using it for simple consultation - and we only do it because unfortunately the search has been ruined. Otherwise most of the folks both internally and externally that I know using these tools would be happy to just go back to google search and SO. Unfortunately that's not an option. Also see if your second argument makes any sense at all. Maybe it comes out of a lacking math background? Firstly, you don't need two zeroes to get a zero out at the end of the multiplication. And secondly, if an average engineer is a zero, what are folks like you then? But again, it maybe just your own projections...
14 replies →
You seem to be assuming that the negative multiplier is on the human side of the equation. There’s your mistake
Alternative hypothesis is that you work on trivial problems, and therefore you get a lot of help from LLMs. Have you considered this?
1 reply →
This is like a person who thinks that making a photocopy of an Einstein paper makes him Einstein. You know, Einstein wasn't that special after all and the photocopier affects his fundamental ego.
I've actually written my own a homebrew framework like this which is a.) cli-coder agnostic and b.) leans heavily on git worktrees [0].
The secret weapon to this approach is asking for 2-4 solutions to your prompt running in parallel. This helps avoid the most time consuming aspect of ai-coding: reviewing a large commit, and ultimately finding the approach to the ai took is hopeless or requires major revision.
By generating multiple solutions, you can cutdown investing fully into the first solution and use clever ways to select from all the 2-4 candidate solutions and usually apply a small tweak at the end. Anyone else doing something like this?
[0]: https://github.com/sutt/agro
There is a related idea called "alloying" where the 2-4 candidate solutions are pursued in parallel with different models, yielding better results vs any single model. Very interesting ideas.
https://xbow.com/blog/alloy-agents
Exactly what I was looking for, thanks.
I've been doing something similiar: aider+gpt-5, claude-code+sonnet, gemini-cli+2.5-pro. I want to coder-cli next.
A main problem with this approach is summarizing the different approaches before drilling down into reviewing the best approach.
Looking at a `git diff --stat` across all the model outputs can give you a good measure of if there was an existing common pattern for your requested implementation. If only one of the models adds code to a module that the others do not, it's usually a good jumping off point to exploring the differing assumptions each of the agents built towards.
This reminds me of an an approach in mcmc where you run mutiple chains at different temperatures and then share the results between them (replica exchange MCMC sampling) the goal being not to get stuck in one “solution”
> "I have more ideas than time to try them out" — The problem we're solving
I see a possible paradox here.
For exploration, my goal is _to learn_. Trying out multiple things is not wasting time, it's an intensive learning experience. It's not about finding what works fast, but understanding why the thing that works best works best. I want to go through it. Maybe that's just me though, and most people just want to get it done quickly.
Yeah, this seems like the opposite of invention. You can throw paint at a canvas but it won’t make you Pollock. And will you feel a sense of accomplishment?
Hey all! I'm one of a handful of developers on this project. Great to see it's getting some interest!
For context, we are right in the middle of building this thing... multiple rebuilds daily since we are using it to build itself. The value isn't in the code itself, yet, but in the approaches (UNIX philosophy, meta-cognitive recipes, etc.)
We are really excited about how productive these approaches are even in this early stage. We are able to have amplifier go off make significant progress unattended for sometimes hours at a time. This, of course, raises a lot of questions on how software will be built in the near future... questions which we are leaning into.
Most of our team's projects, unless they have some unresolved IP or are using internal-only systems, are built in the open. This is a research project at this stage. We recognize this approach it too expensive and too hacky for most independent developers (we're spending thousands of dollars daily on tokens). But once the patterns are identified, we expect we'll all find ways to make them more accessible.
The whole point of this is to experiment and learn fast.
Here's a writeup of the project for more context: https://paradox921.medium.com/amplifier-notes-from-an-experi...
I do a lot of work with claude code and codex cli but frankly as soon as I see all the LLM-tells in the readme, and then all the commit messages written by claude, I immediately don't want to read the readme or try the project until someone else recommends it to me.
This is gaining stars and forks but I don't know if that's just because it's under the github.com/microsoft, and I don't really know how much that means.
Future LLMs are going to be trained on this. Github really ought to start tagging repos that are vibe-coded.
I'd rather have in-depth commit messages then three word ones
When I blind-commit claude code commit messages they are sometimes totally wrong. Not even hallucinations necessarily - by the time I'm committing the context may be large and confusing, or some context lost.
I'd rather have the three word message than detailed but wrong messages.
I think I agree with you anyway on average. Most of the time a claude-authored commit message is better than a garbage message.
But it's still a red flag that the project may be filled with holes and not really ready for other people. It's just so easy to vibe your way to a project that works for you but is buggy and missing tons of features for anyone who strays from your use case.
1 reply →
Contributors
claude Claude
Interesting given Microsoft’s history with OpenAI
History in AI is rewritten on a daily basis
https://techcrunch.com/2025/09/09/microsoft-to-lessen-relian...
more than history -- early, massive investment in OpenAI by Microsoft and formerly their exclusive compute provider.
This stood out to me too, seems like a months-long project with heavy use of Claude
Starting in Claude bypass mode does not give me confidence:
WARNING: Claude Code running in Bypass Permissions mode │ │ │ │ In Bypass Permissions mode, Claude Code will not ask for your approval before running potentially dangerous commands. │ │ This mode should only be used in a sandboxed container/VM that has restricted internet access and can easily be restored if damaged.
The Readme clearly states:
Caution
This project is a research demonstrator. It is in early development and may change significantly. Using permissive AI tools in your repository requires careful attention to security considerations and careful human supervision, and even then things can still go wrong. Use it with caution, and at your own risk.
Claude Code will not ask for your approval before running potentially dangerous commands.
and
requires careful attention to security considerations and careful human supervision
is a bit orthogonal no?
4 replies →
I assumed, especially with the VS Code recommendation, that this would automatically use devcontainers...
If they didn't have this warning you'd see comments on how irresponsible they are being
A lot of the ideas in this aren't bad, but in general it's hacky. Context export? Just use industry standard observability! This is so bad it makes me cringe. Parallel worktrees? These are prone to putting your repo in bad states when you run a lot of agents, and you have to deal with security, just put your agent in a container and have it clone the repo. Everything this project does it's doing the wrong way.
I have a repo that shows you how to do this stuff the correct way that's very easy to adapt, along with a detailed explanation, just do yourself a favor, skip the amateur hour re-implementations and instrument/silo your agents properly: https://sibylline.dev/articles/2025-10-04-hacking-claude-cod...
I'll always be skeptical about using AI to amplify AI. I think humans are needed to amplify AI since humans are so far documented to be significantly more creative and proactive in pushing the frontier than AI. I know, it's maybe a radical concept to digest.
> I'll always be skeptical about using AI to amplify AI.
This project was in part written by Claude, so for better or worse I think we're at least 3 levels deep here (AI-written code which directs an AI to direct other AIs to write code).
I think I'm more optimistic about this than brute-forcing model training with ever larger datasets, myself. Here's why.
Most models I've benchmarked, even the expensive proprietary models, tend to lose coherence when the context grows beyond a certain size. The thing is, they typically do not need the entire context to perform whatever step of the process is currently going on.
And there appears to be a lot of experimentation going on along the line of having subagents in charge of curating the long term view of the context to feed more focused work items to other subagents, and I find that genuinely intriguing.
My hope is that this approach will eventually become refined enough that we'll get dependable capability out of cheap open weight models. That might come in darn handy, depending on the blast radius of the bubble burst.
Based on clear, operational definitions, AI is definitely more creative than humans. E.g., can easily produce higher scores on a Torrance test of divergent thinking. Humans may still be more innovative (defined as creativity adopted into larger systems), though that may be changing.
More creative? I've just seen my premium subscription "AI" struggling to find a trivial issue of a missing import in a very small / toy project. Maybe these tools are getting all sorts of scores on all sorts of benchmarks, I dont doubt it, but why are there no significant real-world results after more than 3 years of hype? It reminds of that situation when the geniuses at Google offered the job to the guy who created Homebrew and then rejected him after he supposedly did not do well on one of those algorithmic tasks (inverting a binary tree? - not sure if I remember correctly). There are also all sorts of people scoring super high on various IQ tests, but what counts, with humans as with the supposed AI is the real world results. Benchmarks without results do not mean anything.
It is as creative as it's training material.
You think it is creative because you lack the knowledge of what it has learnt.
This is absurd to the point of being comical. Do you really believe that?
If an “objective” test purports to show that AI is more creative than humans then I’m sorry but the test is deeply flawed. I don’t even need to look at the methodology to confidently state that.
3 replies →
> Never lose context again. Amplifier automatically exports your entire conversation before compaction, preserving all the details that would otherwise be lost. When Claude Code compacts your conversation to stay within token limits, you can instantly restore the full history.
If this is restoring the entire context (and looking at the source code, it seems like it is just reloading the entire context) how does this not result in an infinite compaction loop?
I think the idea would be that you could re-compact with a different focus. When you compact, you can give Claude instructions on what is important to retain and what can be discarded. If you later discover that actually you wanted something you discarded during a previous compaction, this could allow you to recover it.
Also, it can be useful to compact before it is strictly necessary to compact (before you are at max context length). So there could be a case where you decide you need to "undo" one of these types of early compactions for some reason.
Hi all, I'm the primary author/lead on the "research exploration" that is Amplifier at Microsoft. It's still SUPER early and we're running fast and applying learnings from the past couple of years in new ways to explore some new value we're finding early evidence of. I apologize that the repo is in a very rough condition, we're running very fast and most of what is in there now has been very helpful but will very soon be completely replaced with our next major iteration of it as we continue to run ahead. I did want to take a pause today and put together a blog post to capture a little more context for those of you here who are following along:
https://paradox921.medium.com/amplifier-notes-from-an-experi...
For those who find it useful in this very early stage, to find some value for yourself in either using it or learning from it, happy to be on the journey together. For those who don't like it or don't understand why or what we're doing, I apologize again, it's definitely not for everyone at this stage, if ever, so no offense taken.
Project looks interesting, but no demos. As much I want to try it because of all cool concepts mentioned, but I am not sure I want to invest my time if I don't see any demos
I mean that's fair but doing a make install and providing your API key is pretty easy?
multiply it by 20 other similar projects and assume 20% have security issues, your environment will be messed up before you even understand if you need it or not. Not even talking about time you lost
There are hundreds of these on github. Why should we care? Why not release any benchmarks or examples?
The very first line in the readme is a quote, attributed to "the problem we're solving".
That's cute
If you think about it, that's because "the problem we're solving" is running out of time. Once it's solved it won't be able to try out ideas.
Is this is a Claude Code wrapper?
Yes
Didn’t GitHub create something similar called Spec.
You are thinking of https://github.com/github/spec-kit
Billions in investment into OpenAI and this is a wrapper for Claude API usage. This is very much a microsoft product.
>"Amplifier is a complete development environment that takes AI coding assistants and supercharges them with discovered patterns, specialized expertise, and powerful automation — turning a helpful assistant into a force multiplier that can deliver complex solutions with minimal hand-holding."
Again this "supercharging" nonsense? Maybe in Satiyas confabulated AI-powered universe, but not in the real world I am afraid...
I was hoping this was going to be an awesome new music player, but no, everything new thing is AI now. Welcome to the future
README files in the "ai_context" directory provide the ultimate AI Slop reading experience..
Yeah, I'm not even that opposed to using AI for documentation if it helps, but everything from Microsoft recently has been full-on slop. It's almost like they're trying to make sure you can't miss it's AI generated.
"Eat your own dog slop" isn't bad practice, though.
Some people in the organization will experience the limitations and some will learn — although there are bound to be people elsewhere in the organization who have a vested interest in not learning anything and pushing the product regardless.
[dead]
[flagged]
[flagged]
Ok, but please don't post shallow dismissals of other people's work to HN. This is in the site guidelines: https://news.ycombinator.com/newsguidelines.html.
You're right, my comment was not good.
[flagged]
[flagged]
Sorry, is this Hacker News? This kind of project is exactly what I'd expect hackers to create. Not using AI in boring limited practical ways where it's known to somehow work, but supercharging AI with AI with AI... etc, and seeing what happens!
Sounds like a research project, they're sharing it out to get some feedback and get a discussion going.
How is this different than Google's Jules thing? Both sort of experimental exploratory things.
Why are you doing free research work for a profit making entity. Are you paid for it.
Well, stop asking silly questions. How will the execs get their bonuses, if it turns out we fucked up the web search and invested an equivalent of a moonbase in ... well, I hate to use the phrase, but statistical parrot ?
That's essentially what a CI environment does. "Multiple tabs" and "swarms". This part should feel familiar to any developer. Having multiple things running in the background to help you is not a new concept and we've been doing it for decades.
Whether these new helpers that explore ideas on their own are helpful or not, and for which cases, is another discussion.
[flagged]
[flagged]
Please don't post like this to this site. It's against the guidelines (https://news.ycombinator.com/newsguidelines.html) because we're trying for something else here.
My apologies, won’t happen again
[flagged]
From the Hacker News Guidelines:
"Please don't post comments saying that HN is turning into Reddit. It's a semi-noob illusion, as old as the hills."
https://news.ycombinator.com/newsguidelines.html
It seems like this discussion is full of shallow dismissals and smug takes though. See the sister comment to yours.
1 reply →
[flagged]
[flagged]
I see you're being down voted, Reddit style. But you're on the mark about the hate tone of comments. If you don't like Amplifier, don't use it. No need to spew hate.
[flagged]
No it doesn’t. It’s dead easy to get a decent level, and going further requires individual effort and skill—-just like any other field of endeavor.
Gatekeepers who claim otherwise have something to sell.
How can it be gatekeeping when they are literally making it easier to use? The analogy is probably closer to a Linux distro. You can put everything together yourself but if someone gives you a pre integrated environment with best practices it makes it easier to get started
6 replies →
> A lot of developers either don’t use coding agents to their full potential
Define "full potential".
Sounds like you are just making things up to sell your product.
Our “product” is a tool we developed internally and found it so useful that decided to open source it.
With full potential I refer to getting the best possible results. For example, being able to work on tasks in parallel without Claude instances interfering with each other vs , well, no doing so.
2 replies →
Common etiquette is to declare your conflicts of interest
Agreed, it is in my bio but I updated the post in any case
3 replies →
aka Winamp
Is this going to be another HN dropbox moment?
Can we get Windows 7 back instead? Nadella rode the cloud wave in an easy upmarket, his "AI" obsession will fail. No one wants this.
The Austrian army already switched to LibreOffice for security reasons, we don't need another spyware and code stealing tool.
> No one wants this
There are many many people who want better AI coding tools, myself included. It might or might not fail, but there is a clear and strong opportunity here, that it would be foolish of any large tech company to not pursue.
Their own employees have to be surveilled and coerced to use their own dog food:
https://news.ycombinator.com/item?id=45540174
> Nadella rode the cloud wave in an easy upmarket
I would say it’s more the result of anti competitive bundling of cloud things into existing enterprise contracts rather than the wave. Microsoft is far worse than it ever was in the 90s but there’s no semblance of antitrust action in America.