Research Engineer
At a glance
Mid-level AI Research role at Doji. New York · full-time · no salary stated · we estimate $181,850–$300,000.
Doji has not published a salary for this role. The figure below is our estimate, not theirs.
aimlroles summary, based on the employer's posting.
What you'll do
- Build generative AI features for the product, including avatar and virtual try-on models
- Create pipelines that prepare large image and video collections for model training
- Use user feedback and preference tuning to improve model quality
- Advance personalized try-on for different body types, clothing, styles, video and 3D
- Work in a shared codebase with strong coding practices and partner with the founders
What you bring
- Hands-on experience training and fine-tuning diffusion models is required
- You know how to implement and optimize pipelines for diffusion-model training
- You have built machine-learning papers from scratch using PyTorch or JAX
Who this fits
This is an in-person role at the company’s New York City office. It suits a hands-on research engineer who enjoys working independently, learning deeply and building creative machine-learning products. Experience with virtual try-on or consumer products is a bonus.
From the employer
The Opportunity
We're reimagining fashion shopping with AI avatars. To do this we're developing cutting-edge diffusion models and designing novel, personalized interfaces.
We're a small team of creative builders in NYC with a rare combination of taste and deep AI expertise. Previously we've shipped consumer products to millions at Apple, DeepMind, Meta, and startups. Backed by the investors behind OpenAI, Cursor, and SKIMS, our company is positioned at the intersection of AI, technology, and culture.
The Role
We're looking for Research Engineers to help us build the next generation of our avatar and virtual try-on models. This role is in-person from our office in NYC.
As Research Engineer you will…
Implement cutting-edge generative AI features for inclusion in our product
Build data pipelines to feed large-scale image and video datasets into model training
Enhance model quality through user feedback and preference-based tuning
Improve virtual try-on for diverse body types, clothing styles, and aesthetic treatments
Collaborate effectively in a shared codebase while upholding high coding standards
Develop state-of-the-art personalized try-on capabilities in video and 3D
Work directly with our founders to create industry-leading technology
You're a great fit if you…
Have hands-on experience training and fine-tuning diffusion models (MUST-HAVE requirement)
Are familiar with implementing and optimizing diffusion model training pipelines
Care deeply about craft and creating category-defining experiences
Have worked in rigorous, fast-paced environments (i.e. startups) or have been a very productive IC in academia or open-source
Are self-driven, high-agency, and can learn new things quickly and deeply
Have implemented ML papers from scratch in pytorch or jax
Get excited about new ML methods, and read papers for fun
Have instincts on how to get unstuck when things don't work
Have genuine interest in fashion, avatars, or creative expression
Bonus if you…
Have experience with virtual try on
Have built consumer products