What We Do:
Orchard Robotics builds AI-powered camera systems that collect valuable data for farmers, providing comprehensive insights into their crops across thousands of acres. Their state-of-the-art AI analyzes billions of fruits, delivering accurate yield estimates, fruit counts, size projections, disease detection, inventories, and bloom maps. All data is managed within their cloud platform, FruitScope OS, enabling farmers to make precise crop management decisions.
The Role:
Orchard Robotics is seeking a Machine Learning Engineer to build creative, practical, and robust solutions for ML/CV software and infrastructure problems. This role focuses on training edge ML models using massive amounts of real-world farm image data collected by their camera systems.
About the role:
- Full-time, in-person role at their San Francisco or Seattle office.
- As an early engineer, you'll receive generous equity compensation.
- Comprehensive Health, Vision, and Dental coverage, with 100% of the premium covered.
- Work in a fast-paced environment with a close-knit, highly driven team, directly collaborating with the CEO and entire team.
- Contribute to a mission-driven company making a positive impact by reducing food waste and increasing food supply.
What you’ll do:
- Build and maintain scalable ETL pipelines for processing large, diverse image datasets from tractor-mounted camera systems.
- Develop and deploy infrastructure for model training, evaluation, and inference, for both cloud and edge devices.
- Design and implement intelligent active sampling infrastructure to optimize data collection and improve model performance.
- Stay current with literature in computer vision models and architectures and apply relevant advancements.
- Collaborate with a multidisciplinary team to integrate ML solutions into production robotics systems.
- Work closely with agronomists and farmers to translate crop biology domain knowledge into actionable ML features.
- Act as a generalist, supporting various parts of the software stack as needed.
What makes you a good fit:
- 2+ years of real-world, industry experience building production-grade data pipelines and ML infrastructure.
- Proficiency in Python and experience with ML frameworks (e.g., PyTorch).
- Strong experience with data engineering tools (e.g., Pandas, SQL, MLFlow, WandB).
- Familiarity with cloud platforms (AWS, GCP) and containerization (Docker, Kubernetes).
- Experience working with massive amounts of real-world training data.
- Familiarity with MLops software and data engineering to ensure consistent deployment of ML models.
- Ability to work independently, learn quickly, and operate in a dynamic environment.
- Enthusiasm for taking on multiple roles and responsibilities as the company grows.
Bonus Points:
- Experience deploying & optimizing ML models to run fast on embedded compute like NVIDIA Jetson.
- Experience prototyping, evaluating, or deploying new ML/CV models on the edge.