Member of Technical Staff - Post Training

Black Forest Labs seeks a Senior Member of Technical Staff to own the full post-training pipeline for its generative models, aligning them with human intent and enabling user personalization across text-to-image and video modalities.

This senior individual contributor position focuses on transforming foundation models into production-ready products. The role is central to the post-training phase, where the team aligns models with human preferences, aesthetic judgment, and creative styles across multiple modalities.

Responsibilities

• Manage the entire post-training workflow, including data curation, reward modeling, fine-tuning, preference optimization, distillation, safety adjustments, evaluation, and deployment.
• Develop and improve methods across the post-training stack, such as SFT, RLHF, RLAIF, DPO, and reward modeling, to better align models with human intent and aesthetic standards.
• Apply post-training techniques to various modalities, including text-to-image, image editing, multi-reference, and video generation.
• Create personalization and customization features that allow users to adapt the models to their unique creative preferences.
• Design and maintain high-throughput infrastructure for fine-tuning and evaluation to enable rapid iteration for the research team.
• Identify quality and alignment gaps through rigorous evaluation and address them through targeted research and engineering efforts.

Requirements

• Proven track record of owning the post-training process for a frontier generative model, including SFT, preference optimization (DPO or RLHF), distillation, and safety tuning, with measurable improvements in human preferences or benchmarks.
• Extensive experience across the entire post-training stack, including reward modeling, preference learning, RLHF/RLAIF, and personalization.
• Comfortable working with multiple modalities, such as text-to-image, image editing, multi-reference, and ideally video.
• Strong proficiency in PyTorch, with the ability to write research code that is robust and reusable by others.
• Focus on delivering measurable model-quality improvements that reach end users, rather than just publishing papers.

Nice to have

• Experience with distillation techniques (LADD, DMD, consistency models, or similar) or building high-throughput evaluation pipelines is a strong plus.

What the company offers

• Base annual salary ranging from €130,000 to €340,000, plus equity.
• Coverage of reasonable travel costs to facilitate the role.

About the company

Black Forest Labs is the team behind Latent Diffusion, Stable Diffusion, and FLUX, foundational technologies that have transformed image and video creation. The company operates as a frontier research lab with a strong commitment to research excellence, open science, and expanding human creativity. Headquartered in Freiburg, Germany, with a growing presence in San Francisco, the company values in-person collaboration, meritocratic ideas, and a warm, empathetic culture where people come first.

• The company is known for foundational technologies like Latent Diffusion, Stable Diffusion, and FLUX.
• Models power tools used by millions of creators, developers, and businesses worldwide.
• FLUX is among the most advanced generative systems globally.
• Headquartered in Freiburg, Germany, with a growing presence in San Francisco.
• Values research excellence, open science, and expanding human creativity.
• In-person time is prioritized and made accessible.
• Focus on getting the science right, deep understanding, and beautiful products.
• Meritocratic culture where the best idea wins regardless of who proposed it.
• Shared credit and no hierarchy above any task.
• Ambitious approach with a focus on shipping rather than waiting for perfection.
• People-first culture emphasizing genuine warmth and agency without empathy.