ComfyUI Engineer.
Runflow · United States
- Hybrid
- Full-time
- $150,000 / year
- United States
Tailored resume — keyword-matched to this role.
Hiring manager — we find who's hiring.
Intro email — drafted to reach them directly.
Job highlights
- Own AI image workflows powering millions of images monthly.
- Build and optimize production-ready ComfyUI workflows.
- Reduce VRAM, latency, and costs without quality loss.
- Integrate workflows with APIs and production systems.
- Train LoRAs and fine-tune diffusion models.
About the role
ComfyUI Engineer
Own the workflows that power millions of AI images every month. You think in nodes, you debug in graphs, you care obsessively about output quality.
The Role
Runflow is the AI image and video infrastructure layer for production. We power the AI headshots, fashion shoots, and product photography that companies like BetterPic and BetterStudio ship every day, millions of images a month, with quality scored before delivery. We're hiring a ComfyUI Engineer to own the workflows that produce them. This isn't a research role. You'll ship production workflows that directly impact output quality, customer retention, and revenue. You're the person who decides which checkpoints, which LoRAs, which control nets, which sampler, and how it all stays under VRAM and latency budgets at scale.
If you debug node graphs in your sleep, train LoRAs for fun, and have strong opinions on face similarity vs. likeness preservation, this is for you.
Quick Facts
- Team: Workflows
- Reports to: Head of AI
- Location: Remote (Global)
- Type: Full-time or contract
- Compensation: Senior IC + equity
- Start: ASAP
Representative Projects (First 6 Months)
- Ship a face-similarity workflow that holds 99%+ likeness across 500-image batches.
- Cut p95 latency on the BetterStudio pipeline by 40%.
- Train a domain-specific LoRA from scratch for a new product vertical.
Responsibilities
- Own the full lifecycle, from prototype to production at scale.
- Build, optimize, and debug ComfyUI workflows for human-centric and product image generation.
- Profile node chains. Cut VRAM, latency, and cost without sacrificing quality.
- Wire workflows into Runflow's Solutions API, Python, TypeScript, REST.
- Run batch inference and post-processing at production scale (millions/mo).
- Iterate on output quality. Hand off failure modes to Sentinel; learn from its scores.
- Build dynamic conditioning systems, variable poses, styles, backgrounds, products.
- Maintain a versioned, documented workflow repo other engineers can read.
- Train and fine-tune checkpoints + LoRAs (Flux, SDXL, Wan-style pipelines).
Requirements
- Shipped ComfyUI in production for 6+ months, setup, custom nodes, debugging gnarly graphs.
- Deep working knowledge of diffusion models (SDXL, Flux, the open-source release of the month).
- Built workflows tuned for realism and identity preservation; face similarity is second nature.
- Comfortable with ControlNets, IP-Adapters, segmentation, and the full conditioning toolbox.
- Optimize compute: know when to swap a sampler vs. retrain a LoRA.
- Fluent in GitHub, Python, and REST; the workflow is just one part of a system.
Strong Candidates Also Have:
- Trained LoRAs or fine-tuned checkpoints (Flux/SDXL/Wan) end to end.
- Use the diffusers library and Hugging Face tooling fluently.
- Built integrations connecting ComfyUI to external services (Plytix, Shopify, custom CMSes).
- Shipped production pipelines at scale (>100K images/day) and lived to tell the war stories.
- A public portfolio of workflows you'd be proud to walk us through.
Tech Stack
- Backend: Node.js, Python, PostgreSQL, Redis
- Workflow: ComfyUI, diffusers, PyTorch, Hugging Face
- Infra: AWS, Cloudflare R2, Modal, Neon
- Tools: GitHub, Linear, Slack, Sentinel
How We Hire
Five steps. Decision speed is part of the offer. We move fast because senior candidates with multiple offers reward the team that respects their time. Whole loop fits in two weeks.
- Application Submitted: Form, ~5 min. We read every word of every application. No silent rejections, ever. Triaged within 5 business days.
- Take-home Challenge: 2 hours max, your time. A small, real Runflow problem. We score the prompts and decisions you made as much as the output itself. Reviewed within 3 business days.
- 30 minutes with the CTO: 30 min, live. Quick conversation about your take-home, the team, and how you work day-to-day. Decision in 48 hours.
- Paid Challenge, 2 days: 2 days, we pay for your time. A bigger problem we pay you to work through. We care about outcomes, not process. AI tools more than welcome. Decision in 48 hours.
- Closing Interview: 90 min, live. Casual chat with the two founders. Decision in 24 hours.
Key Skills/Competency
- ComfyUI
- AI Image Generation
- Diffusion Models
- Workflow Optimization
- Python
- LoRA Training
- ControlNet
- IP-Adapter
- Production Scale Inference
- REST APIs
Skills & topics
- ComfyUI Engineer
- ComfyUI
- AI Image Generation
- Diffusion Models
- Workflow Optimization
- Python
- LoRA Training
- ControlNet
- IP-Adapter
- Production Scale Inference
- REST APIs
- Machine Learning
- Remote
- Full-time
- Contract
How to get hired
- Tailor your application: Highlight experience with ComfyUI production, diffusion models, and workflow optimization.
- Showcase your work: Include specific examples of face-similarity workflows, batch generation pipelines, or custom conditioning logic.
- Demonstrate technical skills: Emphasize proficiency in Python, GitHub, REST APIs, and experience with LoRA training or checkpoint fine-tuning.
- Prepare for challenges: Expect a take-home challenge and a paid 2-day challenge focusing on real Runflow problems.
- Articulate your vision: Be ready to discuss your taste in AI tools and demonstrate how you reach for AI as a default solution.
Technical preparation
Master ComfyUI node graph construction and debugging.,Deep dive into diffusion models like SDXL and Flux.,Practice optimizing VRAM and latency in workflows.,Build and integrate workflows using Python and REST.
Behavioral questions
Describe a complex workflow debug you solved.,How do you balance quality with performance?,Share your opinion on a product you find beautiful.,Explain how you reach for AI as a default tool.
Frequently asked questions
- What is Runflow seeking in a ComfyUI Engineer?
- Runflow is seeking a ComfyUI Engineer who can own and optimize production workflows for AI image generation, focusing on output quality, efficiency (VRAM, latency), and integration with their API. They value engineers who can ship end-to-end solutions and have a strong aesthetic sense, rather than specific years of experience or degrees.
- How long does the hiring process take at Runflow for a ComfyUI Engineer?
- Runflow emphasizes speed in its hiring process, aiming to complete the entire loop within two weeks. They provide decisions within 24-48 hours after interviews and challenges, respecting candidates' time, especially those with multiple offers.
- What technical skills are most important for the ComfyUI Engineer role?
- Key technical skills include production experience with ComfyUI (6+ months), deep knowledge of diffusion models (SDXL, Flux), proficiency with conditioning tools like ControlNets and IP-Adapters, Python, GitHub, and REST APIs. Experience in optimizing compute resources and training LoRAs or fine-tuning checkpoints is highly valued.
- Is experience with specific AI tools like LoRAs or ControlNets required for the ComfyUI Engineer position?
- While not strictly required, strong candidates for the ComfyUI Engineer role will have experience with tools like LoRAs and ControlNets, as well as IP-Adapters and segmentation. The job description highlights that deep working knowledge of diffusion models and the ability to build tuned workflows for realism and identity preservation are crucial.
- What kind of projects can I expect to work on as a ComfyUI Engineer at Runflow?
- As a ComfyUI Engineer at Runflow, you'll work on concrete, shippable projects such as developing face-similarity workflows with high likeness preservation, reducing latency in existing pipelines by a significant margin, and training domain-specific LoRAs from scratch to streamline future product vertical onboarding.
- Does Runflow require a specific degree for the ComfyUI Engineer role?
- No, Runflow explicitly states they do not look for a specific number of years, a specific degree, or a specific stack. They hire based on a candidate's ability to ship end-to-end solutions and their 'taste' or aesthetic judgment, viewing other factors as secondary.
- How does Runflow assess candidates' 'taste' for the ComfyUI Engineer role?
- Runflow assesses 'taste' through application questions that prompt candidates to share opinions on beautiful products, discuss AI tools they find effective, and showcase their work. They value genuine opinions and the ability to articulate why certain aesthetics or outputs are appealing, as taste is considered non-negotiable.
- What is the compensation structure for the ComfyUI Engineer position?
- The compensation for the ComfyUI Engineer role is structured as a senior individual contributor (IC) role with the addition of equity. This suggests a competitive salary and a stake in the company's success.
Similar roles
Open positions we recommend based on this role.