Assistant Research Engineer - Computer Vision

Description

The Johns Hopkins Center for Bioengineering Innovation & Design (CBID) in the Department of Biomedical Engineering seeks an Assistant Research Engineer to lead computer vision and AI development for the Vector Cam platform. Vector Cam is an AI-enabled mobile imaging system that enables community health workers to identify mosquito species in real time, supporting faster vector surveillance and improved malaria control strategies.

This role serves as the technical lead on computer vision and image analysis, designing and iterating on machine learning architectures, managing training pipelines and datasets, and optimizing models for deployment across edge and cloud environments. The position requires an experimental, curiosity-driven mindset that continuously explores new model architectures and approaches while pushing performance and reliability, strong attention to data management, and expertise in communicating statistical reasoning behind model choices.

Department:
Johns Hopkins Center for Bioengineering Innovation & Design (CBID), Department of Biomedical Engineering, Whiting School of Engineering

Location:

Baltimore, MD, USA (in person)

Reports to:

Dr. Soumyadipta Acharya (Principal Investigator)

Design, train, and evaluate computer vision models for mosquito identification and related vector‑borne disease projects; develop and maintain scalable training and evaluation pipelines for image classification and detection models; continuously explore and benchmark new architectures, training approaches, and optimization strategies to improve model accuracy and robustness; design and maintain data management systems ensuring clean, versioned, and traceable datasets; develop strategies for deploying models on mobile edge devices and cloud infrastructure, optimizing inference for smartphones and other resource‑constrained platforms;

collaborate with software engineers to integrate models into the Android application and imaging pipeline; investigate and troubleshoot performance issues related to camera systems, imaging conditions, and device variability; develop benchmarking methods to monitor model performance across deployments; collaborate with entomologists and field teams to improve data collection and labeling; contribute to publications and presentations describing algorithm development and system performance.

Computer Vision and Model Development: Design and train deep learning models for insect classification and morphological recognition using architectures such as Efficient Net, YOLO, Vision Transformers, and other modern approaches; devise strategies for limited datasets, noisy data, and challenging real‑world image conditions.

Model Optimization for Edge Deployment: Optimize models for deployment on smartphones (Tensor Flow Lite, PyTorch Mobile, ONNX), investigate quantization and pruning techniques, ensure consistent performance across different smartphone cameras.

AI Data Pipeline and Dataset Management: Build systems for dataset versioning, experiment tracking, and model reproducibility; maintain clear documentation of dataset lineage and experiment configurations; support continuous retraining as new field data becomes available.

System Architecture for AI Deployment: Design architecture for managing model updates, versioning, and deployment across edge devices and cloud platforms; develop monitoring strategies to maintain reliability across large‑scale field deployments.

Project Impact: Vector Cam enables rapid, accurate mosquito species identification in the field, strengthening malaria control programs and supporting targeted interventions.

Master's degree in Computer Science, Machine Learning, Computer Vision, Software Engineering, or a related field.

Strong background in computer vision and deep learning, experience training and evaluating models using PyTorch or Tensor Flow.

Solid understanding of probability, statistics, and model evaluation methods, with the ability to explain model choices and performance metrics.

Experience with image datasets, data pipelines, and model evaluation methodologies; experience…