Introducing Ivy, our first step in connecting AI agents with the real world
Give your claude code / codex the compute, tools, and memory to work with video on a Macbook
Introducing Ivy, our first step in connecting AI agents with the real world
Give your claude code / codex the compute, tools, and memory to work with video on a Macbook
paste the following to your agent
Set up Ivy - https://t.co/N42n5F233O
and you will get a free onboarding and demo straight on your local agent
works in codex, claude code, and any other agent interface with MCP
Annotating and processing 15k episodes (5.7+ days, 200+ tasks) of open-source robot eval datasets containing failures
For safe deployment, we must cheaply understand failures and unexpected behavior
You can't dump terabytes of robotics data onto a laptop and point an agent at it, but thatโs what we do with most smaller shapes of data
We made working with physical data as easy as working with CSV files.
Today, we are launching Interlatent Dataroom
@robotwinnem as models get better and cheaper it will be more cost-efficient than human annotation (for some pipelines it already is), so we're working on the infra as this happens!