Your CEO wants to know what finance is doing with AI.
Your team already uses it. But it stops at the edge of your ERP, bank and portals.
Meet Actions by CoPlane: AI that works inside those systems, with your team approving every change. coplane.com/perspectives/int…
Today @coplane is releasing Localsandbox: a lightweight library to give your agents a sandboxed filesystem with safe bash and code execution.
It's built on top of @tursodatabase's incredible AgentFS project, @vercel's just-bash library, and Pyodide.
Try it out today! 👇
Today we're announcing our $14M seed round led by @RibbitCapital, with participation from @Stripe, @OptumVentures, and @TerrainCap.
Thank you to our investors and customers for trusting us to build the future of the enterprise back office with them.
We're building the best team in enterprise AI and hiring across the board.If you want to build novel but pragmatic AI solutions to genuinely hard problems with $1bn+ revenue customers — we'd love to hear from you.
coplane.com/careers
Bring your SOPs and connect your existing systems.
CoPlane uses your historical data as evals to create self-assembling, self-improving workflows that outperform humans.
Deploy in hours, not years. Your cloud, your data, your controls.
We're building CoPlane to change that.
An AI intelligence layer that works across your existing systems — SAP, Workday, AS/400, Netsuite, Excel, etc — to power agentic workflows that actually adapt to how your business runs.
Enterprise back offices haven't meaningfully improved in decades. G&A as a % of revenue across the S&P 500 has been stuck at ~15% for 30+ years.
The ERP industrial complex — consultants, BPOs, point solutions — charges a fortune and doesn't move the needle.
If you're excited about helping define what the UX and pattern language for a future of agentic and self-assembling enterprise software, please reach out!
We're looking for NYC-based design engineers to join our small, talent-dense team.
This is an opportunity to define the frontier of HCI. We're building the platform for self-assembling, self-optimizing software already in production with consequential enterprises.
Our eng team is made up of folks who have built and maintained large open-source projects like Neovim and Analytics.js and scaled massive data systems at companies like Google, Stripe, Wix, and Segment.
DM if you are interested or know someone who might be!
We're looking for NYC-based design engineers to join our small, talent-dense team.
This is an opportunity to define the frontier of HCI. We're building the platform for self-assembling, self-optimizing software already in production with consequential enterprises.
Our eng team is made up of folks who have built and maintained large open-source projects like Neovim and Analytics.js and scaled massive data systems at companies like Google, Stripe, Wix, and Segment.
DM if you are interested or know someone who might be!
I have a feeling people like the *idea* of creating with node-based canvases more than actually doing it.
They're great for 2 things IMO:
- Visualizing & understanding existing flows
- Refining existing logic
Otherwise, natural text is the most intuitive way to create.
Lack of multi-line lambdas in Python make simple APIs for durable workflows inconvenient. Not sure how to make this work with typing but this is one option I found. Function transforms into its return value.
With all the buzz around @vercel 's `use workflow`, thought I'd share a bit about our pythonic durable workflow framework at @coplane, planar. You can think of it like Django or Rails for agentic apps.
Everyone is finally coming around to the idea that agents should be durably executed but we've been building durable agents with our customers since before it was cool.
We've built an agent framework directly into Planar, so you don't need to compose various libraries together. But planar goes further, out of the box you get human-in-the-loop, files support, a rules engine, and everything else you need to rapidly build durable agent apps.