AI Engineer - Computer Vision

Company Description

Antoniq Robotics is harnessing the power of Physical AI to shape the future and sustainability of agriculture.

We're building one of the most advanced AI and hardware systems on the planet, with embodied AI for agriculture: robots that perceive, decide, and act in demanding, real-world environments.

Our perception runs directly on the robot as it operates in the field — multimodal, real-time, and up against scenes that are small, clustered, occluded, and variable from site to site.

We're looking for an AI Engineer to own the perception pipeline end to end: from how multimodal data is captured in the field, through annotation strategy and deep learning and generative model development, to inference analytics, field validation, and deployment on the robot.

What You'll Be Doing

• Evaluate and adapt the multimodal-capture setup for in-field perception
• Define the annotation schema
• Train and iterate deep learning detection and segmentation models with a focus on small and occluded objects: high-resolution inference, tiling and overlap handling, duplicate control.
• Build the detection and analytics layer
• Define validation criteria and run field tests
• Optimize models for real-time inference on embedded / edge hardware and integrate the AI module on the robot.
• Keep datasets, annotations, model versions, and experiments versioned and reproducible.

What We're Looking For

• Strong Python and a modern deep-learning framework.
• Hands-on object detection and segmentation experience, including training on custom datasets.
• Practical dataset skills: collection protocol design, annotation tooling, schema design, quality control.
• Rigorous evaluation habits: held-out sets, per-error-mode analysis, metrics beyond a single headline number.
• Genuine curiosity about Physical AI. You want your models acting in the real world, not just scoring on a benchmark.
• A pull toward the physical world. You're comfortable collecting data and validating models in agricultural field conditions, and at home in a fast-moving startup.
• Strong plus: real-time / embedded deployment, multi-object tracking across frames, and perception, localization and mapping, and sensor fusion.

Logistics

• Candidates must have the right to work in Europe; no visa sponsorship is available.
• Hybrid / on-site working in Porto, Portugal.

Want to ship computer vision that farmers rely on?

Send your CV to [email protected]