Sandor Felber

@thejesonlee @moribots I know a guy

if you're building full stack robotics for the triple D (Dull, Dirty, Dangerous) and you know hard tech deeply, please DM me.

@jh_naylor @thejesonlee Loading ... ;)

@thejesonlee @SandorFelber

Is anyone building / exploring robotics for dangerous jobs where safety is critical? DM me.

@thejesonlee @MinervaHumanoid Coming out of stealth soon 🦾🦿

@thejesonlee We are at @MinervaHumanoid

@thejesonlee @MinervaHumanoid Come join us in our mission to save human lives in the most dangerous jobs of the planet using humanoids! (Predominantly in energy, power generation, and oil and gas currently)

At Vmax, we are automating the construction of RL environments and the post-training of agents. We are hiring members of technical staff and research fellows. Come join us in SF! (link to apply in comments).

We are getting humanoid robots to do real physical work - not demos - and are looking to meet people in robotics & reinforcement learning! If you’ve shipped robots and know your way around reinforcement learning (locomotion or manipulation), teleoperation, VR, sim-to-real, safety, and latency - message me 5-10 lines about your background and a link to your portfolio/website/GitHub/repo if interested.

Introducing Bunny - world's first curiosity device for kids It’s screenfree..it’s portable.. We raised $1M from @southpkcommons to reimagine how kids thrive in the age of AI, safely. Comment 'Bunny'. Our nephew will pick 50 families that get it for free this holiday season…

Happy to share my O1 application has been approved earlier this summer - shout out to @lisawehden and the amazing @plymouthstreet for compiling a solid application for me so quickly! Excited to get back to building @NyroHumanoids at @southpkcommons in San Francisco!

One week remaining to apply for the Founder Fellowship @southpkcommons Here's why it's the best way to start a venture scale company

What is it like to build humanoid robots, as a software engineer? Today's deepdive goes into the software side of this industry - but before we do, we need to talk about hardware. With Sandor Felber, formerly Humanoid Robot Learning Researcher at MIT: newsletter.pragmaticengineer.com/p/humanoid-rob…

Catch @nolan_fey's oral/poster at today's ICLR Robot Learning Workshop! robot-learning.ml/2025/ The paper will also appear at RSS.

1/ Peter Thiel famously said, "We wanted flying cars, but we got 140 characters instead." I've always wondered: why? Why wouldn't we build flying cars in Silicon Valley? There are lots of reasons...but I think the simple one is: Hardware is slow 👇

Casting reward selection as a model selection leads up to 8x faster learning and 50% better performance! (arxiv.org/abs/2410.13837) ⚡ Provable regret guarantees. 🌟 Easy to implement (github.com/Improbable-AI/…). ⚔️ 1 GPU can do the work of up to 8 GPUs! Presenting ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization to be presented at @iclr_conf 2025. Designing and selecting reward functions is one of the most challenging problems in building AI agents. ORSO accelerates progress in building performant AI agents. Joint work with @calvincbzhang @ZhangWeiHong9 and @aldopacchiano

Excited to share my recent work with @gabe_mrgl , Martin Pettico, and @pulkitology . We’re pushing the limits of whole-body control to make robots faster, stronger, and more athletic!