Member of Technical Staff — Inference-Multimodal & Diffusion

RadixArk; Palo Alto (CA); On-site; USD 200,000 - 400,000; Full time

Job summary

RadixArk is seeking a Member of Technical Staff — Inference-Multimodal & Diffusion to drive the frontier of generative modeling. You will work on diffusion and flow-based models for image, video, and multimodal generation, pushing quality, efficiency, and scalability.

You will blend deep research with engineering execution—designing novel algorithms, training at scale, and deploying models for researchers, developers, and real-world applications.

Qualifications

• 5+ years of experience in ML research or applied ML engineering.
• Strong expertise in diffusion models or generative models (DDPM, DDIM, latent diffusion, flow matching, etc.).
• Deep understanding of deep learning fundamentals and optimization.
• Proven experience training large-scale models on GPUs/TPUs.
• Strong proficiency in PyTorch or JAX.
• Experience implementing research ideas into working systems.
• Strong mathematical foundation in probability, statistics, and optimization.

Responsibilities

• Design and develop next-generation diffusion and generative models.
• Improve model quality, controllability, and sample efficiency.
• Research and implement novel training and sampling methods.
• Optimize models for large-scale distributed training.
• Collaborate with systems teams to scale training and inference.
• Translate research ideas into practical production systems.
• Evaluate models using rigorous metrics and benchmarks.
• Contribute to long-term research and product direction in generative AI.

Skills

ML research/Applied ML engineering

Diffusion models

Deep learning fundamentals

GPUs/TPUs training

PyTorch or JAX

Research-to-production

Probability and optimization

Tools

PyTorch

JAX

Job description

About the Role

RadixArk is seeking aMember of Technical Staff —Inference-Multimodal & Diffusionto advance the frontier of generative modeling.

You will work on cutting-edge diffusion and flow-based models for image, video, and multimodal generation, pushing model quality, efficiency, and scalability. This role combines deep research thinking with strong engineering execution — from designing novel algorithms to training and deploying models at scale.

Your work will directly shape next-generation generative AI systems used by researchers, developers, and real-world applications.

This is a high-impact role for engineers and researchers who want to push the limits of generative models in both theory and practice.

Requirements

• 5+ years of experience in ML research or applied ML engineering
• Strong expertise in diffusion models or generative models (DDPM, DDIM, latent diffusion, flow matching, etc.)
• Deep understanding of deep learning fundamentals and optimization
• Proven experience training large-scale models on GPUs/TPUs
• Strong proficiency in PyTorch or JAX
• Experience implementing research ideas into working systems
• Strong mathematical foundation in probability, statistics, and optimization
• Ability to move from research prototypes to production-quality models

Strong Plus

• Publications in top-tier conferences (NeurIPS, ICML, ICLR, CVPR, etc.)
• Experience with large-scale distributed training
• Experience in multimodal generation (text-to-image, video, audio)
• Familiarity with transformer architectures and hybrid models
• Experience improving sampling speed and generation efficiency
• Contributions to open-source generative model projects
• Experience scaling models to billions of parameters

Responsibilities

• Design and develop next-generation diffusion and generative models
• Improve model quality, controllability, and sample efficiency
• Research and implement novel training and sampling methods
• Optimize models for large-scale distributed training
• Collaborate with systems teams to scale training and inference
• Translate research ideas into practical production systems
• Evaluate models using rigorous metrics and benchmarks
• Contribute to long-term research and product direction in generative AI

About RadixArk

RadixArk is an infrastructure-first company built by engineers who've shipped production AI systems, created SGLang (30K+ GitHub stars, the fastest open LLM serving engine), and developed Miles (our large-scale RL framework). Founded by AI infrastructure veterans from xAI and NVIDIA, we're on a mission to democratize frontier-level AI infrastructure by building world-class open systems for inference and training. Our team has optimized kernels serving billions of tokens daily, designed distributed training systems coordinating 10,000+ GPUs, and contributed to infrastructure that powers leading AI companies and research labs.

Compensation

Depending on background, skills, and experience, the expected annual salary range for this position is $200,000 - $400,000 USD + equity.

Equal Opportunity

RadixArk is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.