6+ years of engineering experience in production ML systems
Hands-on experience with ML model serving and APIs (e.g., PyTorch)
End-to-end experience building applied ML systems with safety components
Solid infrastructure skills in Kubernetes, containerization, and cloud platforms (AWS, GCP, Azure)
Experience collaborating with customer engineering teams in complex environments
Strong Python skills for reading and modifying model code
Ability to communicate technical trade-offs to both technical and non-technical stakeholders
Willingness to take on Technical Deployment Lead responsibilities for strategic accounts
Responsibilities
Build custom endpoints and pipelines on FLUX for customer workflows
Take customer use cases from scoping to production endpoint
Own deployment of FLUX models into customer environments
Profile customer workloads and apply inference optimizations
Build evaluation sets against customer requirements and catch quality regressions
Help customers adapt models through finetuning and training pipelines
Advise on deployment architecture and trade-offs between setups
Feed field learnings back to research and platform teams and create reusable tooling
Benefits
Work in a distributed team with real office space
Join the team in the office at least 2 days a week
Engage in a culture grounded in values of obsession, low ego, boldness, and kindness
Opportunity for occasional national and international travel for team offsites and customer engagements
Full Job Description
What You'll Work On
As a Forward Deployed Engineer, ML Infrastructure & Inference at BFL, you will:
Build custom endpoints and pipelines on top of FLUX for our customers: multi-step workflows, prompt harnesses and upsampling, pre- and post-processing, moderation and safety layers, and combinations of FLUX with VLMs and other models
Take customer use cases from scoping to proof of concept to a production endpoint, running on BFL's or the customer's infrastructure
Own the deployment of FLUX models and custom pipelines into customers' production environments, from on premise to BFL-hosted integrations, including serving setup, GPU sizing and scaling
Profile customer workloads, apply existing inference optimizations to hit cost and latency targets, and partner with BFL's inference team on deeper optimization work
Build evaluation sets and harnesses against customer requirements, and catch quality regressions across model versions
Help customers adapt models to their use case through finetuning and LoRA training pipelines on their own or BFL's infrastructure
Advise customers on deployment architecture and on the trade-offs between BFL-hosted and self-hosted setups (cost, latency, security, compliance)
Feed field learnings back to BFL's research and platform teams, and turn repeated customer work into reusable tooling, reference architectures and documentation
Act as a technical thought leader within the FDE team, setting standards for applied and deployment work and mentoring other engineers
What We're Looking For
6+ years of engineering experience, with significant time building and running ML systems in production
Hands-on experience building and serving ML models in production, from model code to API (e.g. PyTorch, model serving frameworks, GPU workloads)
Experience building applied ML systems end to end: pipelines combining multiple models, evaluation, and safety or moderation components
Solid infrastructure skills: Kubernetes, containerization, observability, and at least one major cloud platform (AWS, GCP, Azure)
Experience working directly with customer engineering teams in complex enterprise environments, including security reviews and regulated or air-gapped setups
Strong Python skills and comfort reading and modifying model code
Ability to explain complex technical trade-offs clearly to both engineers and non-technical stakeholders, internally and externally
Willingness to take on Technical Deployment Lead (TDL) responsibility for strategic accounts, owning the technical relationship and delivery end to end
Willingness to travel occasionally, nationally and internationally, for team offsites and customer engagements
Nice to have
Prior knowledge of diffusion models and/or flow matching, and relevant finetuning and distillation techniques
Inference optimization experience and a track record of improving latency, throughput or cost per generation
Experience with LLMs or VLMs for prompt harnesses, captioning or evaluation
Experience with open source tooling from the FLUX ecosystem such as ComfyUI, diffusers and common trainer frameworks
Experience serving video generation models
Open-source contributions to inference, serving or diffusion model projects
Previous experience as a forward deployed engineer, solutions architect or ML platform engineer at an AI lab or cloud provider
How We Work Together
We're a distributed team with real offices that people actually use. This role is based in San Francisco, and you'll join us in the office at least 2 days a week, or one full week every other week.
Everything we do is grounded in four values:
Obsessed. We are a frontier research lab. The science has to be right, the understanding deep, the product beautiful.
Low Ego. The work speaks. The best idea wins, no matter who said it. Credit is shared. Nobody is above any task.
Bold. We take the ambitious bet. We ship, we do not wait for conditions to be perfect.
Kind. People over politics. We treat each other with genuine warmth. Agency without empathy creates chaos.