OpenHands
@OpenHandsDev
OpenHands is the leading open source agent for software development, usable through a CLI, GUI, SDK, or IDE
23 Following 11.6K Followers
AI is becoming a system you build, not just a model you call.
Models, inference, agents, evals, infrastructure. Builders should be able to choose each layer and change it as better options emerge.
That shared philosophy is why OpenHands is excited to be a partner for the @NebiusAI Builder Program.
Show more
The August ClickPy report is out: overall downloads cooled over summer time, PyPI recorded 176.6B downloads in August 2026, down 4.5% from July.
But MCP and AI-agent tooling are still accelerating. 🤖
Notable packages this month:
🔌 mcp-types: 11.7M → 61.3M (+425%), MCP tooling just became a 60M-download dependency
☁️ google-cloud-parametermanager: 862K → 4.82M (+459%), @googlecloud, a 5.6x jump for Google Cloud’s Parameter Manager client
🎨 skia-python: 167K → 4.02M (+2,312%), Python bindings for Google’s graphics engine suddenly went 24x
🤝 openhands: 1.11M → 2.52M (+128%), @OpenHandsDev, the autonomous coding agent has now doubled for four straight months
📈 metricflow: 4.31M → 8.11M (+88%), @dbt_labs, semantic-layer adoption keeps compounding
Full report here:
Show more
Agentic development is moving beyond one-off chats.
The useful part isn’t just getting an agent to complete a task once. It’s being able to understand what it did, review the output, and turn that work into something repeatable.
That's where we focused in August across eight Agent Canvas releases:
⚡ Better automations with live run status, dashboards, Git sync, and reusable configuration
🎛️ More control over models, providers, context, cost, and agent behavior
🧩 Canvas Apps beta, plus better ways to inspect files, commits, and agent output
Read the full August product update:
Show more
I like how simple the @OpenHandsDev memory approach is for their software factory.
Agents just write to a project-level MEMORY.md file.
This is powerful in three ways:
1. It allows agent workers to leave behind state and be ephemeral.
2. It creates a basic shared-learning mechanism. If one agent learns something useful, every future agent can benefit.
3. Each agent is expected to be a good steward for the greater good, in this case the project, and not just think about its own task.
Show more
We're happy to be an OSS partner with the new Stanford class on modern software development. We're looking forward to lots of contributions!
I’m excited to finally announce the newest edition my Stanford course 𝗧𝗵𝗲 𝗠𝗼𝗱𝗲𝗿𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿. It has been 9 months in the making.
Last November, with the release of Claude Opus 4.5, coding agents experienced a step function improvement in capability. We all felt it. The LLMs were more powerful, could reason for longer, solve harder tasks.
This year’s iteration of my course reflects the 2026 metamorphosis of software engineering.
My core belief is simple: AI-native developers of the LLM era are going to become the most important members of any software organization. I have designed my course to train this next generation of engineers.
𝗪𝗵𝗮𝘁’𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝘁𝗵𝗶𝘀 𝘁𝗶𝗺𝗲 𝗮𝗿𝗼𝘂𝗻𝗱
First, 85% of my Fall 2025 class material is being thrown out. The Fall 2026 syllabus reflects the core capabilities AI-native engineers must have: agent skills, advanced context engineering, MCP portals, agent-ready codebase principles, agentic code review, security, parallelizing background agents, software factories, and more.
Second, I am going to teach my students how to have software taste. Every student will be required to ship pull requests to production-grade, real-world codebases. The course is collaborating with the top open-source AI repos who will offer support and mentorship to students on how to meaningfully contribute to their projects.
This has never been done before in any university course so I am incredibly grateful to our OSS Partners: @browserbase, @HeyGen, @CopilotKit, @semgrep, @OpenHandsDev, @milvusio, @marimo_io, Pi, @crewAIInc, @warpdotdev, @vercel, @cmux, @arizeai, @UnslothAI, and @anyscalecompute.
𝗪𝗵𝗮𝘁’𝘀 𝘀𝘁𝗮𝘆𝗶𝗻𝗴 𝘁𝗵𝗲 𝘀𝗮𝗺𝗲
I’m fortunate to again have AI software engineering leaders and founders as guest speakers to share their learnings from building top coding agent products. Thank you to @leerob from @cursor_ai, @bcherny of @claudeai code, @EnoReyes of @FactoryAI, @silasalberti of @cognition, @0xine of @semgrep, Rajesh Bhatia of @Cloudflare , @amasad of @Replit, and @eladgil.
All resources will be available online. All classes will be available to the public.
9/22 on Stanford campus. See you in class.
Show more
Better models do not eliminate the need for good agent design.
At ODSC AI West 2026, @rajistics, Agentic AI Engineer at @OpenHandsDev, will lead the training “The Design of Coding Agents: A Hands-On Workshop on Harness Engineering.”
Rajiv will break down four core harness levers that shape coding-agent performance: retrieval, memory, loops, and architecture. Attendees will test these ideas live using the open-source OpenHands harness and see how different design choices affect agent behavior.
With experience across OpenHands, Contextual AI, Hugging Face, Snorkel, Snowflake, and DataRobot, Rajiv brings a practical view of what it takes to build coding agents that work beyond a demo.
👉 Join this AI for Software Developers training at ODSC AI West 2026:
Show more
AI sprawl gets expensive fast. Not because teams are using too much AI, but because agents, models, and workflows start spreading across the org with no shared way to measure what’s working.
The answer isn’t forcing everyone onto one model. It’s building shared infrastructure to:
• route work to the right model
• compare cost against outcomes
• reuse successful workflows
• turn one-off agent wins into automations
@rbren_dev spoke with @usatoday about what it takes to move from scattered AI tools to agent infrastructure that engineering teams can actually operate at scale.
Read the full article:
Show more
You iPad is now a work station: Install @OpenHandsDev on @umbrel, access through @Tailscale and view file changes in the browser-rendered VSCode 🤯
There are SO MANY great models that you can run in coding agents for free or virtually for free now.
- Ox Alpha, @Zai_org's new GLM Flash model, free for a little longer:
- Inkling by @thinkymachines is free on OpenRouter:
- Qwen-3.8-27B can be run on any decent sized hardware at home:
- deepseek-v4-flash by @deepseek_ai is a great model that is very cheap, we provide it free on the OpenHands Cloud:
If you're not trying these you should be!
Show more
Interviewing our wonderful panel from last night!
Thank you to all who showed up and listened in. We enjoyed hosting @nvidia, @OpenHandsDev, and @ollama at our SF office. Stay tuned for the next one! 👀
Show more
This week with OpenHands 👋
Kimi K3 is the new free default, plus updates to Automations, conversation context, first-run guidance, LLM provider credentials, and Jira Cloud setup.
Show more
A developer prompting their way through a Jira ticket gets about 20 to 30% faster, says @rbren_dev. So that's why he says the real gains sit somewhere else.
On the new episode of First Commit, Laura Hamilton sits down with the co-founder and CEO of @OpenHandsDev, the open source project that became one of the most starred AI coding agent repos on GitHub.
Show more
I love how agents make it possible to just automate little random things in your life. Right now I have agents:
1. watching my training runs and fixing them when they break
2. preemptively fixing broken CI and merge conflicts
3. preparing a daily work plan for me at 6AM
Show more
The next productivity leap in software engineering won’t come from engineers manually prompting more coding agents.
It comes when the work starts itself.
PR opens → agent reviews it
Datadog alert fires → agent investigates
Coverage drops → agent adds tests
Developers shift from initiating every task to supervising always-on agent workflows.
@rbren_dev joined @ProductFaculty to show how we’re building toward that at OpenHands. Watch the episode:
Show more
This Thursday, August 20th, we’re bringing together four leaders at the forefront of open models, coding agents, and AI infrastructure:
Kyle Kranen, Senior Manager, Dynamo at @nvidia
Yunmo Koo, Founding Engineer at @friendliai
Jeffrey Morgan, CEO at @ollama
Saurya Velagapudi, Principal Engineer at @OpenHandsDev
Together, they’ll explore what it takes to move AI coding agents to production and how open models are changing the economics of AI-native engineering.
Join us for the conversation, followed by audience Q&A and networking. Spots limited!
📍 San Francisco
📅 Aug. 20
🎟️ RSVP →
Show more
@OpenHandsDev, @ollama, and @friendliai take on whether open models — Kimi, GLM, MiniMax — have closed the gap enough to run production coding agents. One moderated panel, 30 minutes of open Q&A, and networking over food and drinks.
If you own the budget, the headcount, or the vendor strategy, this conversation is for you.
Grab your spot while there's still room!
Aug 20, 6 PM →
Show more
AI security improves when the infrastructure around the model is as open as the model itself.
Once agents can access repos, execute commands, use credentials, and call tools, security has to extend across identity, permissions, sandboxes, policies, logs, and execution.
We’re excited to join the Open Secure AI Alliance alongside @NVIDIA and other industry leaders working together, in the open, to advance AI security across the whole stack.
Show more
A new DeepSeek V4 Pro is out and we're offering it for free for a limited time on OpenHands Cloud!
Pop over to to sign up, and select deepseek-v4-pro.
We've been using deepseek-v4-flash a lot, so pro will certainly be a step up!
Show more
We’re proud to be an @NVIDIAAI launch partner and excited to support NVIDIA Nemotron 3.5 Lightning! ⚡
Always-on agents need fast, customizable models, but they also need infrastructure: harnesses, runtimes, sandboxes, integrations, policies, audit trails, and deployment flexibility. That’s where OpenHands comes in.
Learn how to use Nemotron 3.5 Lightning & OpenHands:
Show more
One thing I've done a lot recently is have agents organize my daily work, especially my todo list and communication channels (e.g. slack/email).
I find it super-useful when there's a bunch of stuff to keep track of, so I created a video here:
Show more
