Computer Vision Engineer - Intern

Title: Computer Vision Engineer - Intern

Vistras Technologies, is a deep-tech startup building an enterprise-grade platform for critical asset monitoring and early anomaly detection. We leverage live thermal imaging, AI, and computer vision to predict and prevent failures in industrial settings, from energy storage facilities to manufacturing plants. We are now expanding our product umbrella to include advanced process monitoring and video analytics, and we need your help to build it.

Job Description:
As a Computer Vision Intern, you will be integral to our R&D team, working directly with the CTO to design, develop, and deploy novel computer vision models. Your primary focus will be on extracting actionable intelligence from thermal and visual-spectrum video streams. You will tackle real-world problems, and your code will directly impact the capabilities of our product.

Key Responsibilities:

• Research & Development: Explore and implement state-of-the-art computer vision algorithms for object detection, segmentation, and activity recognition in both thermal and RGB video feeds.
• Model Development: Build, train, and optimize models for specific use cases, such as: Anomaly Detection: Identifying unusual thermal patterns or visual behaviors that precede equipment failure. Process Monitoring: Verifying correct assembly line operations, counting objects, or detecting operational deviations. Multi-Modal Fusion: Intelligently combining data from thermal and visual cameras to improve detection accuracy and robustness.
• Pipeline Implementation: Develop and optimize the end-to-end computer vision pipeline, from data preprocessing and augmentation to model inference on edge devices or cloud platforms.
• Performance Optimization: Work on model quantization, pruning, and acceleration to ensure real-time performance in resource-constrained environments.
• Data Curation: Assist in labeling, cleaning, and managing our proprietary dataset of thermal and visual video sequences from industrial environments.

Qualifications:

Must-Haves:

1. Strong proficiency in Python.
2. Hands-on experience with deep learning frameworks such as PyTorch and TensorFlow.
3. Solid understanding of core computer vision concepts: CNNs, R-CNN/YOLO, Transformers, and OpenCV.
4. Experience with one of the following: video analytics, multi-object tracking, or time-series analysis of video data.
5. Experience in conducting multi-faceted research for AI pipeline building.

Nice-to-Haves (You will stand out if you have):

1. Familiarity with thermal imaging data and its unique characteristics (e.g., non-RGB data, temperature-value interpretation).
2. Experience with model deployment (e.g., using TensorRT, ONNX, or OpenVINO).
3. Knowledge of cloud platforms like AWS (Kinesis, SageMaker) or GCP for running large-scale training or inference.
4. Understanding of industrial processes or IoT systems.
5. A portfolio of projects (GitHub) demonstrating your skills.

Why Join Vistras Technologies:

• Be a part of a cutting-edge startup that is transforming the global supply chain industry using AI and blockchain technology.
• Gain hands-on experience working on real-world projects in a dynamic and collaborative environment.
• Develop your skills and knowledge with access to training opportunities, mentorship programs, and resources.
• Work with a team of highly skilled professionals who are passionate about making a positive impact on the world.