2
points
rbren
Joined 265 karma
- If you want to hack on an emerging agent SDK (specializing in software development), join us at https://github.com/OpenHands/software-agent-sdk
- Spoiler: it's not actually that easy. Compaction, security, sandboxing, planning, custom tools--all this is really hard to get right.
We're about to launch an SDK that gives devs all these building blocks, specifically oriented around software agents. Would love feedback if anyone wants to look: https://github.com/OpenHands/software-agent-sdk
- If you're looking for open source agents, which can run locally, in Docker, or in the cloud, and which have a consistent track record of acing benchmark scores like SWE-bench, check out https://github.com/All-Hands-AI/OpenHands
We're about to release our Agent SDK (https://github.com/All-Hands-AI/agent-sdk/) which provides devs with all the nuts and bolts you need to define custom prompts, tools, security profiles, and multi-agent interfaces
- If you're interested in running agents (specifically for software dev) inside a sandbox, OpenHands [1] runs in Docker by default, and can run on Kubernetes or on a raw VM. It gets access to the standard tools (e.g. file ops, bash) as well as a web browser and a Jupyter notebook.
- If you’re looking for a non-slop OSS agent, check out OpenHands: https://github.com/All-Hands-AI/OpenHands?tab=readme-ov-file
(It’s written like 30% of its own code, so maybe a little slop if I’m being fair. But it’s mostly humans! Almost 400 of us now!)
- If you’re looking for an OSS alternative check out OpenHands CLI: https://github.com/All-Hands-AI/OpenHands?tab=readme-ov-file
- Having started OpenDevin (now OpenHands [1]) I can say it's definitely worth renaming. It's very limiting attaching your branding to someone else's
- Quick plug for the OpenHands CLI: https://docs.all-hands.dev/usage/how-to/cli-mode
We're working on creating an SDK that will allow other folks to build their own CLIs with OpenHands, so you can take advantage of our SOTA agent, but implement the TUI/GUI of your dreams.
- Glad to see everyone centering on using OpenHands [1] as the scaffold! Nothing more frustrating than seeing "private scaffold" on a public benchmark report.
- The idea of swarming multiple agents on a task isn't new. And honestly we haven't seen it really work in practice. We've tested multi-agent systems a bunch with OpenHands [1] and have never really seen a bump on benchmark scores despite the massive increase in complexity. There's nothing that many different agents can do that a single generalist can't accomplish on its own.
That said, they can potentially get you a speedup if you have a neatly separable task, and can parallelize the work. But it doesn't lead to some quantum leap in what agents are able to accomplish unsupervised.
I do think some form of multi-agent workflow is going to become important over the next few years, but more because it fits our mental model of the world rather than being some big technological unlock.
- If you're looking for a fully open source, LLM-agnostic alternative to Claude Code and Gemini CLI, check out OpenHands: https://docs.all-hands.dev/usage/how-to/cli-mode
- I’m biased [0], but I think we should be scripting around LLM-agnostic open source agents. This technology is changing software development at its foundations—-we need to ensure we continue to control how we work.
- 2 points
- When Devin launched ~1 year ago, the first thing I did was google "Is Devin open source?" When I saw it wasn't I immediately started working on OpenHands (fka OpenDevin)
At this point, it's very clear our jobs are changing. It's super important to me that we, the engineers, have a say in how that change happens.
What's exciting is that the community seems to agree. We're now one of the top 50 python projects of all time [1], and a huge number of codegen and AI researchers have been contributing to help make our agent the top performer on SWE-bench Verified [2]
We've still got a lot of work ahead of us, but I'm very excited about what the future of software development looks like.
[1] https://github.com/EvanLi/Github-Ranking/blob/master/Top100/... [2] https://www.swebench.com/
- 11 points
- If you're interested in hacking on agent loops, come join us in the OpenHands community!
Here's our (slightly more complicated) agent loop: https://github.com/All-Hands-AI/OpenHands/blob/f7cb2d0f64666...
- When we first started OpenHands (fka OpenDevin) [1], AI-generated PRs would get opened with OpenHands as the PR creator/owner. This created two serious problems:
* First, the person who triggered the AI could approve and merge the PR. No second set of human eyes needed. Essentially bypassed the code review process
* Second, the PR had no clear owner. Many of them would just languish with no one advocating for them to get merged. Worse, if one did get merged and caused problems, there was no one you could hold responsible.
We quickly switched strategies--every PR is owned by a human being. You can still see which _commits_ were done by OpenHands, but your face is on the PR, so you're responsible for it.
- 3 points
- This is a thoughtful article, but I very much disagree with the author's conclusion. (I'm biased though: I'm a co-creator of OpenHands, fka OpenDevin [1])
To be a bit hyperbolic, this is like saying all SaaS companies are just "compute wrappers", and are dead because AWS and GCP can see all their data and do all the same things.
I like to say LLMs are like engines, and we're tasked with building a car. So much goes into crafting a safe, comfortable, efficient end-user experience, and all that sits outside the core competence of companies that are great at training LLMs.
And there are 1000s of different personas, use cases, and workflows to optimize for. This is not a winner-take-all space.
Furthermore, the models themselves are commoditizing quickly. They can be easily swapped out for one another, so apps built on top of LLMs aren't ever beholden to a single model provider.
I'm super excited to have an ecosystem with thousands of LLM-powered apps. We're already starting to see it materialize, and I'm psyched to be part of it.
- Agreed! The comparison is great for estimating the scope of the tasks they're capable of--they do very well with bite-sized tasks that can be individually verified. But their world knowledge is that of a principal engineer!
I think this is why people struggle so much with agents--they see the agent perform magic, then assume it can be trusted with a larger task, where it completely falls down.
- The bigger win comes not from saving keystrokes, but from saving you from a context switch.
Merge conflicts are probably the biggest one for me. I put up a PR and move onto a new task. Someone approves, but now there are conflicts. I could switch off my task, spend 5-10 min remembering the intent of this PR and fixing the issues. Or I could just say "@openhands fix the merge conflicts" and move back to my new task.
- I'm one of the creators of OpenHands (fka OpenDevin). I agree with most of what's been said here, wrt to software agents in general.
We are not even close to the point where AI can "replace" a software engineer. Their code still needs to be reviewed and tested, at least as much as you'd scrutinize the code of a brand new engineer just out of boot camp. I've talked to companies who went all in on AI engineers, only to realize two months later that their codebase was rotting because no one was reviewing the changes.
But once you develop some intuition for how to use them, software agents can be a _massive_ boost to productivity. ~20% of the commits to the OpenHands codebase are now authored or co-authored by OpenHands itself. I especially love asking it to do simple, tedious things like fixing merge conflicts or failing linters. It's great at getting an existing PR over the line.
It's also important to keep in mind that these agents are literally improving on a _weekly_ basis. A few weeks ago we were at the top of the SWE-bench leaderboard; now there are half a dozen agents that have pulled ahead of us. And we're one launch away from leapfrogging back to the top. Exciting times!
MIT license and model agnostic
I’d also keep a close eye on Toad which is launching this month:
https://willmcgugan.github.io/announcing-toad/