Ability Robotics Hiring! Full Time Robotics Model Engineer in - Ricebowl

About Us:

Phridom AI - Ability Robotics is an early-stage robotics and AI company building the core systems behind embodied intelligence. Our technologies help robots understand the physical world, learn intelligent behaviours, and operate reliably in real environments. We’  re still in stealth, but we’  re already working with a select group of partners to validate our systems in high-impact robotics applications. If you're excited to build foundational technologies for the next generation of robotics and want meaningful ownership over hard technical problems, we'd love to hear from you.

Role Overview

Develop, train, evaluate, and deploy learning-based models that enable robots to perceive, decide, and act. You will support the model team across experimentation, dataset use, training infrastructure, evaluation, and integration with real robotic systems.

What you'll work on

· Implement and train models for robot perception, representation learning, planning, policy learning, or multimodal reasoning.

· Build reproducible experiment configurations, training pipelines, checkpoints, and evaluation reports.

· Analyze dataset composition, failure cases, generalization, and model behavior.

· Design metrics and benchmarks that reflect real task success rather than proxy accuracy alone.

· Run ablations and controlled experiments to guide technical decisions.

· Optimize models for inference latency, memory, stability, and hardware constraints.

· Integrate trained models with simulation, data, and robot runtime systems.

· Maintain high-quality research code and convert successful experiments into dependable components.

· Track relevant research and reproduce promising methods when useful.

· Collaborate with data, CV, controls, simulation, and platform engineers through deployment and validation.

Technical Requirements

Must have

· Bachelor’s or higher degree in machine learning, computer science, robotics, or related field.

· At least 5 years of working experience with strong Python and PyTorch skills.

· Hands-on experience training and evaluating deep-learning models.

· Good understanding of optimization, generalization, data splits, metrics, and experiment design.

· Experience with GPU training, checkpointing, logging, and reproducible environments.

· Ability to read research papers and implement methods accurately.

· Strong debugging skills across data, training, and inference.

· Clear communication and disciplined documentation.

Nice to have

· Experience with robot learning, imitation learning, reinforcement learning, vision-language-action models, or diffusion policies.

· Experience with distributed training or large-scale evaluation.

· Experience deploying models on robots or edge devices.

· Publications or meaningful open-source contributions.

How to Apply

Please send your résumé/CV and a short note explaining your relevant experience to ************* Use the subject: “Application — Robotics Model Engineer”. Links to portfolios, publications, GitHub repositories, or representative projects are welcome where relevant.

Benefits and Compensation

We offer a competitive compensation package based on role scope, experience, interview performance, and work location, together with the benefits required by the employing entity. Final details will be discussed with shortlisted candidates.