Senior Machine Learning Engineer

Arcade•San Francisco, CA

2d•Hybrid

About The Position

We’re looking for a product-minded machine learning engineer who’s excited to push the boundaries of what’s possible with AI. At Arcade, you’ll experiment with the latest models to generate visuals, motion, and language that capture brand personality — crafting the foundation for content that builds and improves itself. This is a highly collaborative and high-impact role. You’ll partner closely with design, engineering, and product to build and customize machine learning models. Expect to move fast, own ambitious projects, and see your ideas shape the way people tell product stories.

Requirements

Strong Python + ML engineering background (PyTorch, diffusers, transformers, or similar).
Minimum 5 years experience with hands-on experience fine-tuning or training open-source models.
Deep familiarity with modern inference and training tooling (LoRA, PEFT, DeepSpeed, Ray, Hugging Face, etc.).
Portfolio or prior work demonstrating creative use of generative models (visuals, videos, text, etc.).
Comfort operating across research, engineering, and product — shipping real features, not just experiments.
Located in or willing to relocate to the Bay Area; occasional in-person collaboration required.

Nice To Haves

Familiarity with TypeScript/Next.js and integrating ML APIs into frontend products.

Responsibilities

Own the ML stack end-to-end: Evaluate hosted models, fine-tune open-source ones, and build scalable pipelines for image, video, and text generation.
Develop brand-aligned generation systems: Create techniques for tone, color, and style fidelity across mediums.
Improve generation quality: Build internal evaluation frameworks to measure realism, brand match, and creative diversity.
Prototype and ship: Rapidly turn research ideas into productized APIs that power Arcade’s content generation features.
Collaborate cross-functionally: Work with design and marketing to define what “on-brand” means, and with engineers to integrate ML systems into production.
Lay the foundation for content optimization: Build data pipelines connecting engagement metrics to model tuning.