Forward Deployed Engineer at Reactor Inc

Reactor is building the inference platform for real-time world models. Our goal is to democratize access to real-time intelligence to unlock next-generation use cases in media, physical AI and omni-directional models. We already run the largest and fastest catalog of world models on the market, and work closely with all major model labs to run and optimize their world models, including Alibaba, Amazon, and NVIDIA. Our team is made of alumni from leading firms such as Apple, Meta, Amazon, Microsoft, Luma AI, Adobe, Stability AI, and more. We've raised $74M led by Lightspeed, with Amplify Partners, WNDR, NVIDIA, Sapphire, and other Tier 1 investors participating. Jeffrey Katzenberg, co-founder of DreamWorks Animation, sits on our board of directors.

We're hiring our first Forward Deployed Engineer. You'll be the engineer who gets a lab's world model running on Reactor at production quality, often from inside their office. You own the deployment end to end, then bring what you learned back into the platform.

This isn't a support or demo role. You'll write production code, make real technical and product calls, and work directly with the founders. You'll also get to define what forward deployment looks like here.

Responsibilities

Be the technical owner for new lab partners and customers, from the first onboarding call to production.
Travel to world model labs, integrate their models onto Reactor's infrastructure, and get them running at production quality.
Profile and optimize inference to hit the latency and throughput targets that make real-time interactive video feel real.
Debug end to end, across model code, inference pipelines, serving and infrastructure.
Build proofs of concept and lightweight integrations when that's what unblocks a deployment.
Turn what you hear from partners into clear specs and priorities for our product and infra teams.
Spot what repeats across deployments and turn it into reusable platform features, tooling and docs.

Skills

3+ years as an engineer shipping production code, including real time working directly with external teams or customers.
Production-level Python. You can trace an inference pipeline, find the bottleneck and fix it.
Strong PyTorch and modern ML inference optimization.
You write clearly. You can turn a messy partner call into a spec another engineer can build from.
High ownership. When a partner is blocked, it's your problem until it isn't.
Comfortable with ambiguity, travel, and getting on a plane to unblock a partner.

Nice to Have

Experience with diffusion models, video generation, or world models
Real-time streaming systems
CUDA kernel optimization, TensorRT, or custom inference frameworks
Kubernetes, distributed systems, or cloud-native deployment
Observability tooling (logs, metrics, tracing)
Existing relationships in the generative video or foundation model ecosystem

Benefits

Competitive SF salary and meaningful early equity
Visa sponsorship and relocation support
Generous health, dental, and vision coverage
Travel covered for on-site integrations