Computer Vision Engineering Intern (Academic Year 2026–2027)

About the position Vectech builds AI-powered tools that identify mosquitoes and ticks from images, helping public health organizations make faster, smarter decisions about vector control. Our computer vision models are deployed in the real world, across diverse geographic regions with local species variants and unfamiliar phenotypes. Understanding how that diversity affects model behavior is a key challenge in maintaining reliable deployed systems. This internship continues work initiated over the summer on dataset drift and model behavior in real-world deployments. You'll focus on analyzing factors indicative of dataset drift, and on understanding how our models behave when faced with largely unlabeled data — specimen images without expert identification. This is a part-time role designed to fit around an academic schedule, with meaningful research contributions expected throughout the year. Responsibilities • Collaborate with the product management team to develop and maintain dashboards that track product usage and performance • Analyze data from computer vision systems deployed in the field to generate insights and identify trends • Create clean, intuitive visualizations to support internal reporting and external communications • Assist in preparing figures and data summaries for presentations, publications, or stakeholder meetings • Help ensure data quality and consistency across different sources and platforms • Respond to data requests with curiosity, efficiency, and just enough skepticism to keep us all honest • Participate in team meetings, contribute to project planning, and ask great questions • Make SQL queries, prepare datasets, and train computer vision models on various research problems • Support development of improved modules in Vectech's computer vision suite Requirements • Proficiency in Python, and familiarity with either TensorFlow or PyTorch • Some hands-on experience in computer vision, image processing, or machine learning — research, grad school, coursework, or personal projects all count • Exposure to Linux systems • Cumulative GPA of 2.5 or above • Ability to manage time independently in a part-time remote role alongside academic commitments • Readiness to learn fast and adapt • Genuine curiosity about public health and environmental science Nice-to-haves • Experience with MLOps concepts (model monitoring, dataset drift, retraining pipelines) • Familiarity with SQL and data visualization tools • Experience contributing to shared codebases (Git) • Prior experience analyzing real-world deployed model performance