Robotics AI Research Engineer

Maven Robotics is seeking AI researchers and engineers to build the neural foundations of its general-purpose robots. The role focuses on designing embodied AI models and learning systems that enable robots to reason, adapt, and perform real-world tasks.

Responsibilities:

• Develop, fine-tune, and optimize foundation models for the robot, leveraging advanced techniques such as pretraining, fine-tuning, and retrieval-augmented generation (RAG)
• Translate high-level application goals into foundational AI solutions, delivering robust and scalable models that perform effectively in real-world scenarios
• Build pipelines for training, evaluation, and deployment of models on various platforms, ensuring high performance and adaptability across diverse use cases
• Stay at the forefront of research, incorporating cutting-edge advancements in AI and foundation models to enhance system capabilities and push technical boundaries
• Own critical projects end-to-end, from model design and experimentation to deployment and continuous improvement
• Collaborate with cross-functional teams, including data, software, and systems engineers, to integrate models seamlessly into production workflows
• Architect Embodied Foundations: Design model architectures (VLA, World Models, etc.) that achieve ultra-high task success rates with minimal human demonstrations
• Master Data Efficiency: Develop novel co-training strategies and efficient learning algorithms that leverage diverse data sources—from Internet-scale video to sparse, high-fidelity human interventions
• Generalize Cross Embodiments: Build models capable of zero-shot or few-shot adaptation to new robot configurations, maintaining a high success rate even when proprietary hardware and actuation systems evolve
• Innovate Real-World RL: Formulate and deploy novel Reinforcement Learning and policy extraction methods specifically designed for physical, real-world manipulation
• Design the Data Loop: Collaborate on advanced data collection systems to capture critical human intervention data for model bootstrapping