Fascinating concept, you essentially frame the backend not as a static codebase, but as an adaptive organism that evolves based on real-time usage.
A few things that come to my mind if I were to build this:
The 'Agent-User' Paradox: To make this work, you'd need the initial agents (the ones responding and testing the goals) to be 'chaotic' enough to explore edge cases, but 'structured' enough to provide meaningful feedback to the 'Architect' agent.
The Schema Contract: How would you ensure that as the backend "writes itself," it doesn't break the contract with the frontend? You’d almost need a JIT Documentation layer that updates in lockstep.
Verification: I wonder if the server should run the 'JIT-ed' code in a sandbox first, using the telemetry to verify the goal was met before promoting the code to the main branch.
It’s a massive shift from Code as an Asset to Code as a Runtime Behavior. Have you thought about how you'd handle state/database migrations in a world where the backend is rewriting itself on the fly? It feels to me that you're almost building a lovable for backend services. I've seen a few OS projects like this (e.g. MotiaDev) But none has executed this perfectly yet.
Well on the bright side it's a complete mask off moment for the tech community. I think it is good for these people to expose themselves to the public. They will show you who they really are if you let them.
“If the broad light of day could be let in upon men’s actions, it would purify them as the sun disinfects”. -- Louis Brandeis
Hot take: what has democracy done for us lately? Besides re-electing Donald Trump?
If something can't go on forever, it will eventually stop. That applies to any system that gives stupid people the same political voice as the rest of the electorate. I mean, it seems kind of obvious, doesn't it?
And why did they give up their agency to these shadowy media oligarchs...?
Answer: because they're stupid.
The ones who weren't stupid were impossible to herd to the polls, or at least a lot more difficult. As a result they were outnumbered. Any system that removes the influences you cite will leave the same stupid voters in place, ready to fall for the next con man who comes along.
The problem isn't the money. The problem is the power. I'm tired of giving stupid people so much power over my life.
>why did they give up their agency to these shadowy media oligarchs
decades tearing down education is paying its dues. Once again, from the people who are making you feel like democracy isn't working.
>I'm tired of giving stupid people so much power over my life.
If power is money, boomers still have a lot of power. And they leveraged politics their whole lives to benefit them (even if destruction of the younger generation is a side effect)
If power is votes, then millenials should be the bloc in charge now... but we still had worse turnout than boomers. That really says something.
I disagree that education had any meaningful part to play. It's true that less-educated people were more inclined to vote for Trump, but it's also true that we got all the "education" about Trump between 2016 and 2020 that anyone should have needed.
You can fix ignorance with education, but you can't fix stupidity.
>what has democracy done for us lately? Besides re-electing Donald Trump?
You have the order backwards. This is their exact strategy; spend decades breaking government then have the breakers say "see? government is broken!". Lack of functioning democracy got us Donald Trump.
Why do you think this year they are so gung ho on trying to disenfranchise voters for a midterm? Because they know they are cooked if democracy starts to work again. If nothing else, Mamdami's 6 weeks show exactly how a "government that works for its people" can work. Let's keep pushing for that.
People extracting value from labor to enrich themselves at the expense of society and then using those riches to further corrupt society, to the point where a few dudes own most of the country does not align with my politics.
That's why I'm a socialist and I would invite anyone who thinks things might not be going in the right direction to consider that as well.
I hate that their politic's explicit goal is to make my, yours, and everyone else's lives actively worse. Government is not a business.
We're not getting better healthcare, more and better jobs, more efficient transportation, better city infrastructure, nor more houses. We aren't even getting the cool things shown in cyberpunk dystopias. Hell, we can't even ask for them to follow the law these days.
Why would I want to support them getting into politics? There's a difference betweeen them having different thoughts on how fund, say, self drving cars (which I'm not a fan of) and then all of that above.
I'd prefer to see more of them do so, personally. That said, to watch Tan wading into a local fistfight about school curriculum and housing zoning and whatnot in the age of ICE abduction, targetted political prosecution and wanton macroeconomic regulatory chaos seems... frustrating.
I mean, I kinda agree with him about most of the centrist stuff. But really, Gary? This is what you need to be spending your money and time on?
I wouldn't even be surprised at this point. Seems like the files will unironically unravel 90% of the elite class at this rate.
What an absolute pathetic hill to die on. You have the riches to fund entire industries, explore the whole world or even beyond, to please any hedonistic pleasure you have... but many chose to do one of the 3 unthinkable things in modern humanity.
I dunno about "demand". But sure, I think most of us would prefer people prioritize like we do. And in particular many of us would view this kind of hobby project as tone deaf and tell people about that in public.
My freedom to tell Tan (or you) that he's being an idiot stems from the same place as his freedom to spend his own money on what he wants.
It's nerfed to a point that it feels more like lawyer than a coding assistant now. We were arguing about an 3rd party API ToU for 1 hour last night. VSC Copilot executed it within 1 minute.
To expand on the "Self-Healing" architecture mentioned in point #2:
The hardest mental shift for us was treating Exceptions as Observations. In a standard Python script, a FileNotFoundError is a crash. In Hive, we catch that stack trace, serialize it, and feed it back into the Context Window as a new prompt: "I tried to read the file and failed with this error. Why? And what is the alternative?"
The agent then enters a Reflection Step (e.g., "I might be in the wrong directory, let me run ls first"), generates new code, and retries.
We found this loop alone solved about 70% of the "brittleness" issues we faced in our ERP production environment. The trade-off, of course, is latency and token cost.
I'm curious how others are handling non-deterministic failures in long-running agent pipelines? Are you using simple retries, voting ensembles, or human-in-the-loop?
100% usability right now is not optimal. I'm releasing a new version of the documentation and installation process today. Sync and pull tomorrow, it should be a lot easier.
I really appreciate your efforts!There're so many valuable use cases that people can deliver via your CRM integration (e.g. leads automation, qualification, scoring, engagement, reporting, etc) It'd be great to test your HubSpot integration and see if these use cases can be brought to life.
That's indeed very hard - we're building runtime captures and they are fed to the coding agent (claude code, cursor, etc) to update the agents' codes. However, the runtime data needs to be structured in a certain way so the coding agents won't get confused. We're testing this a lot right now.
Interesting application - how'd you like to implement the credit-risk logics? Do you want to write SQL expressions, mathematical models via Python, or some other way? By design, the framework should containerize these logics. But it'd better learn more.
I grew up helping my dad run his factory so I'm very familiar with ERP systems for manufacturing. A few years ago, when I decided to build a startup and I indentified the biggest problems with ERP is the fact they all just serve as data integration and system of records now - there're not enough processes automation. Therefore, I thought it'd be very meaningful to leverage AI to automate business processes such as PO, Price Requisition, and invoices, etc.
3 years in, I realized that every customer in our space (construction and contracting) want process automation, however, AI is simply not good enough - it's too slow, unpredictable, inconsistent, and overall hard to count on. For example, automating a quote by asking AI to insert dynamic variables from a relational database is hit or miss. Asking voice AI to provide training does not capture full breadth of the offline activities. Asking AI to fill out a work order creates a ton of errors.
Later, we decided that though LLM and the foundation models were progressing fast, the dev tools were lacking way behind, particular behind all the hypes and promises these AI applications claimed. The agents are not reliable, consistent, intelligent, evolving, and chances are the market would demand more apps to keep the party going.
Therefore, we went full open-source. The mission we have in mind is really to "generate reliable agents that can run business processes autonomously". We see all this hype about general computer use (GCUs) and can't help but making an opposing argument - that the AI agents need guardrails, more defined paths, and most importantly consistent results just like a human would need
- Proactive Reasoning (anticipating future needs or consequences)
The project seems to have gained a bit of a traction so far and I hope that you can fork it and tell the community what's missing and what we should be working on. I deeply thank you because the it's truly painful to build and deploy these one-off agents that don't get utilized. (https://github.com/adenhq/hive).
A few things that come to my mind if I were to build this:
The 'Agent-User' Paradox: To make this work, you'd need the initial agents (the ones responding and testing the goals) to be 'chaotic' enough to explore edge cases, but 'structured' enough to provide meaningful feedback to the 'Architect' agent.
The Schema Contract: How would you ensure that as the backend "writes itself," it doesn't break the contract with the frontend? You’d almost need a JIT Documentation layer that updates in lockstep.
Verification: I wonder if the server should run the 'JIT-ed' code in a sandbox first, using the telemetry to verify the goal was met before promoting the code to the main branch.
It’s a massive shift from Code as an Asset to Code as a Runtime Behavior. Have you thought about how you'd handle state/database migrations in a world where the backend is rewriting itself on the fly? It feels to me that you're almost building a lovable for backend services. I've seen a few OS projects like this (e.g. MotiaDev) But none has executed this perfectly yet.
reply