Flexion Robotics FDE

VLA/WAM Integration & Deployment Engineer

  • Location San Francisco, California, United States
  • Seniority mid
  • Posted 2026-11-05

Original posting ↗ You apply on the company site — we never collect applications.

Extracted automatically from public job postings. Always verify details on the original posting before applying.

role details

Deploy and integrate VLA/WAM models on humanoid robots, optimizing inference and integrating with higher-level agents.

Summary generated by AI from the original posting.

Hard requirements to check first

  • Clearance:not mentioned in the posting
  • Work auth:not mentioned in the posting

Skills

PythonPyTorchAgentsVLAVLMdiffusionflow-matchingvideo/world-modelmultimodal architectures

Excerpt from the original posting

About Flexion 
At Flexion, we're building the intelligence layer powering the next generation of humanoid robots. Our mission is to accelerate the transition from fragile prototypes to real-world humanoid deployment. We are founded by leading scientists in robot reinforcement learning (ex-Nvidia, ex-ETH Zürich), and backed by leading international VC firms. In just months, we’ve gone from our first line of code to deploying real humanoid capabilities.

The role 
As our VLA/WAM Integration and Deployment specialist you'll turn our VLA and World-Action Model research into reliable capabilities running on real humanoid robots.
You will work across the full deployment loop: integrating foundation models with agentic systems, optimizing inference, collecting targeted real-world data, running principled evaluations, and feeding deployment experience back into training.
This is a broad, hands-on role for someone who enjoys making advanced models work as part of a complete autonomous system rather than optimizing a single research component in isolation.

Key responsibilities
- VLA/WAM Deployment: Deploy and validate learned policies across multiple humanoid platforms and real-world tasks.
- Fast Inference: Optimize model serving, latency, action generation, asynchronous execution, and closed-loop control for real-time robot operation.
- Agentic Integration: Connect VLA/WAM policies with higher-level agents for task decomposition, policy selection, structured conditioning, verification, recovery, and replanning.
- In-Situ Data Collection: Develop improved workflows for teleoperation, interventions, autonomous rollouts, failures, recoveries, and targeted post-training data collection.
- Evaluation & Testing: Develop reproducible real-robot evaluations, regression suites, failure analysis, and principled comparisons between model versions.
- Deployment-to-Training Loop: Turn real-world successes and failures into datasets and experiments that improve subsequent pre-training and post-training.

- Strong Python and PyTorch skills and experience working with modern deep-learning systems.
- Experience deploying learning-based models in robotics, autonomous systems, or another real-time embodied setting.
- Familiarity with VLA, VLM, diffusion, flow-matching, video/world-model, or related multimodal architectures.
- Strong systems intuition around inference latency, model serving, data pipelines, logging, and debugging complex ML systems.
- Experience designing rigorous e…

→ Flexion Robotics · Workable

Verify before you apply

  • The title, level and compensation match the original posting.
  • The role is still open — postings often close without notice, we drop them within 10 days of disappearing.
  • What "remote" means here: sometimes it is remote within one country only. The location field shows the employer’s own wording.
  • Work authorization, visa sponsorship and security clearance — these are the most common reasons an application goes nowhere.
  • Before you apply