haykmartiros

Riffusion (gen music) | san francisco | in person | 200-300 + equity

training frontier music generation models and building the most powerful and fun music creation product in the world

looking for a world-class infra person

{ hayk at riffusion dot com }

Riffusion (AI Music) | San Francisco (in person) | $200-300k+ equity

We're building the best music creation and listening experiences ever made, by making music interactive in new ways. We're a small team of world class engineers, researchers, and musicians working out of Hayes Valley. It's hard to believe we get paid for what we do.

Looking for:

* ML Research scientists and engineers who are deep in large-scale generative models and love music

* Devops expert deep on multi-node training and infiniband (will pay well for a short term contract)

Message me: { careers @ riffusion.com }

Longer story: The web demo is a way to earn your curiosity. The Riffusion crew is released our free iOS app today, and that's the real juice. We’re thinking constantly about how to create a new instrument with gen music, and for us this is wildly fun experience.

At the core, we’ve been training music foundation models that are high quality, diverse, controllable, and an order of magnitude faster than anything done before. This required a ton of deep investment across the whole stack from compact representations to architectures to deployment infra. The result is fast enough to enable totally new workflows for this stuff.

The primary input is to capture/upload a photo or video with optional personalization, then the app will generate original lyrics and music based on the context. The next time you see your dog or kid do something funny, you can turn that into a vibe in seconds. We’ve been having a blast creating music in moments that we never would have before.

The most remarkable part is the editing, as sampled in the web demo. In the edit view, you can go deep adding multiple prompts on a timeline and controlling their strengths and interactions. You can mix genres into something totally new or make transitions like going from an operatic aria to death metal. The near-real time editing flow is something that to me feels totally new, like trying GarageBand when it first came out.

There’s a ton more we’re excited to launch in the coming weeks as we work to create more musicians in the world. Excited to hear what you all create!

One of the authors here. Short story is, check out our web demo (on desktop Chrome) for real-time music generation and editing! Music is made on the fly by our diffusion models as you control the transition between two random genres. Every time you make a change, you're creating 30s of 44kHz stereo in 0.3s on a GPU via diffusion. Genre transitions are one of the most fun use cases we've found.

Riffusion | Generative Music | San Francisco, CA

Looking for world class ML research engineers and scientists to join our core team building the future of music. If you love creativity, large scale diffusion models, fast infrastructure, and early startup energy, our team may be a fit. We're building something new to inspire hundreds of millions of people. Competitive comp, in-person culture.

Contact: hayk @ riffusion dot com

Skydio - Autonomy Infrastructure | Bay Area - San Mateo | Python & C++

Skydio is the leading US drone company and the world leader in aerial autonomy. Our mission is to make the world more productive, creative, and safe with autonomous flight. Today tens of thousands of our robots are flown at scale to inspect critical infrastructure and provide life-saving intelligence, and we have one of the best robotics teams on the planet in support of these goals.

Our X10 drone features NVIDIA Orin and Qualcomm 865 processors with 85 TOPS of compute capability, along with onboard 360 trinocular camera coverage. We develop real-time 3D reconstruction and semantic AI so that our robot can understand the world around it and make intelligent decisions.

We’re looking for infrastructure engineers and a director for our autonomy team who love making code for flying robots and for making our engineering team more productive. Experience with Python and C++, performance optimization, systems infrastructure, first-principles thinking, 3D visualization, communication and logging. full-stack web apps, and build systems all welcome. Very strong software engineering and computer science fundamentals required.

I lead the autonomy team and you can reach me at { hayk at skydio dot com }.

I

Skydio - Autonomy Infrastructure | Bay Area - San Mateo | Python & C++

Skydio is the leading US drone company and the world leader in aerial autonomy. Our mission is to make the world more productive, creative, and safe with autonomous flight. Today tens of thousands of our robots are flown at scale to inspect critical infrastructure and provide life-saving intelligence, and we have one of the best robotics teams on the planet in support of these goals.

Our X10 drone features NVIDIA Orin and Qualcomm 865 processors with 85 TOPS of compute capability, along with onboard 360 trinocular camera coverage. We develop real-time 3D reconstruction and semantic AI so that our robot can understand the world around it and make intelligent decisions.

We’re looking for infrastructure engineers for our autonomy team who love making code for flying robots and for making our engineering team more productive. Experience with Python and C++, performance optimization, systems infrastructure, first-principles thinking, 3D visualization, communication and logging. full-stack web apps, and build systems all welcome.

I lead the autonomy team and you can reach me at { hayk at skydio dot com }.

as of a few minutes ago, you can mark individual riffs as public or unlisted, in addition to the global setting!

Remix is using latent space interpolation under the hood to generate three variations of a given riff with (light, medium and heavy) modifications

This is flat out not true. Generative AI is meant to approximate the distribution. It’s biased from the data, yes, but if it was just the median it would only spit out one face, one gender, etc.