Teams at Palona AI

Conduct robotics and Physical AI research connecting interaction understanding with real-world robotic systems. Build closed-loop perception, planning, execution, monitoring, and replanning systems; integrate and debug robots; design experiments in restaurant and healthcare environments; evaluate robustness; and contribute to publications, prototypes, and potential patents.

Develop computer vision and vision-language models for image and video understanding in restaurant environments. Build datasets, training strategies, evaluations, and benchmarks; diagnose model failures; and deploy efficient, reliable inference pipelines. The role combines research and engineering across scene understanding, object tracking, activity recognition, temporal reasoning, visual grounding, and multimodal models, while partnering with product and infrastructure teams to bring research into production.

Own model selection, evaluation, and deployment for voice and multimodal restaurant AI agents. Build datasets, experiments, and monitoring to improve accuracy, safety, latency, and cost. Partner with product and engineering to ship model changes with guardrails, rollouts, and clear metrics tied to business outcomes.