At Agwa, we build autonomous hydroponic growth cabinets that grow fresh greens anywhere. Each unit is packed with cameras and sensors, connected to the cloud, and guided by our Virtual Agronomist: an AI system that monitors plant health, makes growing decisions and alerts when something goes wrong.

What you'll do

• Own the Virtual Agronomist's algorithmic core, from architecture and model choice through deployment and field performance

• Analyze complex problems end to end and build a clear work plan and roadmap toward a complete solution

• Develop and improve CNN-based models (classification, detection, segmentation) that assess plant state from images captured in real-world conditions

• Build the data flywheel with our agronomy team: labelling processes, dataset quality, human-in-the-loop decisions, error tracking and retraining

• Combine imaging with sensor data to detect anomalies and support automated decisions

• Deploy algorithms both in the cloud and locally on edge devices

• Work closely with software, hardware, agronomy and product in a small, multidisciplinary R&D team

• Evaluate where LLMs and VLMs can add real value alongside classic algorithms

What we're looking for

• B.Sc. in Electrical Engineering, Physics, Computer Science or a related field

• 6+ years of hands-on experience in computer vision and algorithm development

• Proven track record of taking deep learning models (CNNs) from research to production

• Strong background in image processing and classic algorithms

• Creative thinker, able to design complex logic to solve complex, real-world problems

• Experience developing and deploying algorithms in cloud environments, on IoT products

• Experience running algorithms locally on edge devices in a Linux environment

• Experience working on multidisciplinary products combining hardware, software and real-world data

• Strong Python skills and hands-on experience with modern deep learning frameworks

Big advantages

• M.Sc. or higher

• Background in electro-optics: cameras, sensors, optics and ISP

• Experience with AWS

• Experience with edge inference optimization

Location: Bar Lev Industrial Park (North, between Akko and Carmiel), 4 days in the office + 1 day from home.