Senior Perception Engineer

AgZen•Somerville, MA

2d•Onsite

About The Position

AgZen is a fast-growing precision agriculture company headquartered in Somerville, Massachusetts, built on MIT research. Its RealCoverage product measures and controls droplet coverage at the leaf level, providing growers with real-time visibility into spray performance and reducing chemical and water use by up to 50% without sacrificing yield. AgZen is a small, technically deep team working across fluid mechanics, computer vision, AI, and agricultural environments. The Senior Perception Engineer will contribute to AgZen’s sensing and software portfolio and own and develop the perception platform that helps growers make spraying more efficient. The role is part of the measurement focus group and is based in Somerville, MA, with in-person work required. Success requires communication, flexibility, teamwork, and willingness to take on and own different responsibilities. The engineer will develop and deploy computer vision and deep-learning capabilities for real-time detection and decision-making in spraying systems, including models for segmentation, classification, optical flow, and 3D reconstruction. The role also involves designing end-to-end models and pipelines, enabling efficient edge execution, analyzing performance in the field, and collaborating with software, hardware, firmware, and product teams to improve the system. The engineer will establish model lifecycle practices and communicate progress, risks, and recommendations to engineering and product leadership.

Requirements

Bachelor’s or graduate degree in Computer Science, Electrical Engineering, or a closely related field.
At least 5 years of experience solving computer vision or perception problems, with a foundation in deep learning and machine learning.
Experience with real-time or edge-based inference and integrating models with hardware or embedded systems.
Ability to design experiments, analyze results, and iterate quickly based on field and product requirements.
Deep experience training and validating computer vision models, such as object detection, segmentation, and classification, for robust real-world performance.
Ability to architect end-to-end computer vision pipelines and decompose complex problems across multiple models and stages.
Expertise in MLOps practices and tooling, such as MLFlow, W&B, or similar experiment-tracking and deployment frameworks, for managing the model lifecycle from research to production at scale.
Advanced knowledge of software engineering principles, including software design, source control management, build processes, code reviews, and testing methods.
Advanced scientific Python, including NumPy, Pandas, and scikit-learn.
Hands-on experience with deep-learning frameworks including PyTorch and/or TensorFlow.
Knowledge of current perception and machine-learning research, demonstrated through research-community activity such as reading conference papers or publications.
Experience operating end-to-end data or ML pipelines for reliability, scale, and observability.
Clear communication skills and ability to work effectively with cross-functional teams in a fast-moving environment.

Nice To Haves

Master’s or Ph.D. degree, or equivalent experience, in Computer Science, Robotics, or a related field.
Experience in agriculture or related fields such as environmental or life sciences.
Robust SQL skills and comfort examining data distributions, feature health, and model behavior.
Experience with data science based on real-world physical sensor data.
Experience creating intuitive data visualization tools that make complex data approachable for non-technical users.
Prior experience developing machine-learning models relevant to biological or crop-protection outcomes.
Familiarity with embedded systems and optimizing models for edge devices.

Responsibilities

Develop advanced computer vision models for real-time detection and decision-making in AgZen’s spraying systems.
Implement deep-learning models for object segmentation, classification, optical flow, and 3D reconstruction.
Architect and build end-to-end models, potentially combining existing networks to improve performance.
Use techniques such as quantization to enable power-efficient model execution on edge devices.
Develop methods to improve model accuracy and inference speed in edge deployment scenarios.
Collaborate with software engineering to architect and develop data-processing and annotation pipelines for continuous system improvement.
Work with hardware, firmware, and product teams to integrate vision models with system components.
Analyze field performance, identify root causes of model behavior, propose improvements, and validate updates through experimentation.
Establish best practices for versioning, testing, deployment, and continuous improvement of computer vision models.
Communicate progress, risks, and recommendations clearly and in a structured manner to engineering and product leadership.
Explore and apply advances in CNNs, transformers, and model quantization.