Efference

Embedded compute devices for embodied vision and inference

[Efference @EfferenceAI]

We’re building robotic cameras that treat vision as a software problem, not a hardware problem — just like humans do.

Demos + model performance: https://t.co/8Si3X0FZ4Y

Pre-orders + Specification: https://t.co/EVAu55ULqM

[GEOFF WOO @geoffreywoo]

36 hours of the best of Silicon Valley in less than 2 minutes. Brought to you by @Antifund @jakepaul and me.

[Efference @EfferenceAI]

We are hiring embedded firmware engineers, full stack engineers, and robot learning interns.

If you or anyone you know is interested in robotic perception, please reach out to us.

https://t.co/MD8vzutZ4T

[Gianluca @gianlucabencomo]

Reliable self-driving for hundreds of $$ (cameras) instead of thousands (LiDAR).

We’ve been testing depth maps and object detection at 50m and beyond.

Videos show @EfferenceAI with D435i mounted on the front, side, and rear while driving around Russian Hill / Pacific Heights

[Gianluca @gianlucabencomo]

These are point clouds from: https://t.co/y2PrlCiuQC

D405 cameras are EVERYWHERE in robotics (commonly wrist mounted like what I have on my YAM arms) but depth is rarely used because it’s not very good

Our models make training 3D diffusion policies possible W/O the use of LiDAR

[Y Combinator @ycombinator]

Efference (@EfferenceAI) is building robotic cameras that treat vision as a software problem, not a hardware problem -- just like humans do.

https://t.co/VbPwapjO8I

Congrats on the launch, @gianlucabencomo!

[Neil Nie @neil_nie_]

Depth sensing is a crucial part of any modern robotics stack, but off the shelf stereo cameras rarely deliver the quality and reliability that we need. What @gianlucabencomo is building here is very exciting: a new depth camera with custom, real-time models baked in. He just opened pre-orders. Worth a look if you care about perception and robotics!

[Gianluca @gianlucabencomo]

@EfferenceAI is building robotic cameras because nothing on the market does the things that make 3D vision efficient and effective in humans. Data representations matter (https://t.co/e21Sdm8qgm) and humans have a lot of important design features (https://t.co/1K1QWZTlUn).

A rich and scalable representation for visuomotor policy learning will come from hardware that is inexpensive and computes high-quality geometric information directly on the PCB, which is exactly what we are building.