Rohit Malhotra

Openhands Maintainer | Ex-CTO @sitewizai (acq. by GAIN) | NLP @ CMU | Interested in Agents

[OpenHands @OpenHandsDev]

Chat works well when you’re working with one agent.

As agent usage grows, we need interfaces built around managing the work, not managing individual conversations.

See how OpenHands uses a custom Kanban board to orchestrate agent work: https://t.co/t9uT6LvbWP

[Rohit Malhotra @rohit_malh5]

RT @OpenHandsDev: Open models matter. But securing AI requires openness beyond the model.

Once agents can access repos, execute commands,…

[Andrew Kuncevich @AndrewK404]

I have to point out just how good @OpenHandsDev is

For frontier models, a 5–10% performance drop when switching to another harness is considered normal. Labs build harnesses specifically for their own models
(Though, to be fair, they optimize them across a broader range of domains)

And yet OpenHands beats the model-native harnesses across every model family—and by a wide margin

Right now, it is clearly the highest-quality harness for coding tasks

[Nathan Lambert @natolambert]

I've been getting a lot more hate than usual as I try to speak my mind about regulatory capture / unintentional attacks on open-source.

It's pretty sad, as there are few people in AI that can speak their mind (most companies say they cannot) and I know many people agree with me silently.

I also get people saying that you only say that because it supports the outcomes you want, in a weirdly derogatory way. Of course this is true, but I'm choosing to turn down meaningful wealth so I CAN fight for these values, working at non profits to speak my mind. Building a future that is more inclusive, diverse in the application of AI, and fairer for our children.

I may not always be right, but it has been clear to me for a while that more openness right now will help way more than supporting the closed causes. I continue to re-visit this and don't think everything should be open like some of the open-source absolutists.

I also don't like a lot of my comrades making fun of anthropic, calling the people there evil, etc. Those are not the case.

Trying to stay the course!

[Rohit Malhotra @rohit_malh5]

@valeriechen_ So awesome, congratulations and best of luck!

[Xingyao Wang @xingyaow_]

People have been asking why OpenHands V1 goes the opposite direction from Claude Managed Agents. I finally found the time to write it up.

The production pain points are real, but changing the hosted agent architecture won't solve them for free:
https://t.co/12VOaZor2j

[Graham Neubig @gneubig]

Just released a new frontend for building with agents called the agent canvas! It is:

- Agent agnostic: you can run OpenHands, CC, or Codex
- You can run agents on local, remote, or cloud
- You can set up agent automations on a schedule

It's so much fun to use!

[OpenHands @OpenHandsDev]

PR review on a schedule.
Issue triage every morning.
Repo summary every Monday.
Your agent can do all of it. It just needed somewhere to live.
Introducing Agent Canvas 🙌
https://t.co/A99YRvhZiG

[Rohit Malhotra @rohit_malh5]

Our civilization is based on science and technology.

Its future hinges on our ability to understand, independently verify, and access the technology without the decree of a select few.

https://t.co/xzSiMDO097

[elie @eliebakouch]

mythos will be bad ON PURPOSE on ai "frontier llm research" tasks, this is very very sad for the research community

also the fact that this is un purpose not visible to the user is crazy

[Graham Neubig @gneubig]

First they came for the model builders...

I feel we're getting a glimpse of a future where AI is only provided to a privileged few, and that's not a future I want to live in.

[MiniMax (official) @MiniMax_AI]

Free agentic coding with MiniMax M2.7 on @OpenHandsDev?

Yes, please! Available for a limited time ⚡👀

[Rohit Malhotra @rohit_malh5]

Token efficiency and supreme flexibility is now enabled through LLM profiles!

Personally very excited to switch over to a cheaper model for more repetitive and straightforward tasks

[OpenHands @OpenHandsDev]

We have also partnered with @MiniMax_AI to provide

*free access to agents with MiniMax-M2.7*

for a limited time!

This is a great opportunity to test out how to mix in cost-effective models into your coding workflows.

Try it out here: https://t.co/FQw0hqxzLX

[Zora Wang @ZhiruoW]

Excited to announce our tutorial: "Future of Work in the Age of LLMs" at #ACL2026 in San Diego, July 2! 🌴
There's a lot of speculation about AI and the future of human work. Our tutorial unpacks it from four angles:
→ The landscape of human work
→ How to build LLMs to augment real-world workflows
→ How to evaluate these LLMs
→ The future of work with LLMs/LLM-based agents

[OpenHands @OpenHandsDev]

The most extensive independent benchmark of LLMs for software engineering just got a big update!

- How does GPT-5.5 compare to Opus 4.7?
- Are open models catching up, and in what areas?
- How do cost and performance stack up?

[Harrison Kinsley @Sentdex]

@c0nsumption_ I may. I live on starlink so i might API test it first to see if its worth the download.

Rn im a big fan of qwen3 next 80B 4bit, running that with @OpenHandsDev cli. Qwen models overall are such bangers.

Still using opus 4.6 tho a lot. Need to get myself totally local one day

[OpenHands @OpenHandsDev]

The OpenHands community slack with 5000+ members was suddenly deleted... We will try to restore access.

@SlackHQ please help!

[Robert Brennan @rbren_dev]

@SlackHQ just deleted our community without notice.

Can anyone help us figure out what's going on?

[Robert Brennan @rbren_dev]

Exciting to see OpenHands SDK taking off like this

We've already hit 3M lifetime downloads, more than half of which were in the last 30d.

[Xuhui Zhou @nlpxuhui]

Creating user simulators is a key to evaluating and training models for user-facing agentic applications. But are stronger LLMs better user simulators?

TL;DR: not really.

We ran the largest sim2real study for AI agents to date: 31 LLM simulators vs. 451 real humans across 165 tasks.

Here's what we found (co-lead with @sunweiwei12).

[OpenHands @OpenHandsDev]

Velocity is dead. If AI can generate a compiler, “write more code faster” isn’t the constraint anymore.

Our Chief Architect, Ray Myers, on what matters next: reliability, constraints, and agent systems.
https://t.co/PUkOzgSkMp