Computer Vision Research Engineer - Intern

The role

You will build and improve computer vision models for detection, segmentation, or OCR on construction drawings. You will work on real production problems and contribute to data pipelines, training runs, and evaluation.

What you will do

• Build and improve CV models for detection, segmentation, or OCR on construction drawings
• Contribute to data preparation, labeling workflows, training runs, and evaluation
• Run experiments, measure results, and communicate results
• Dig into real customer data to find failure modes and propose fixes
• Work with fullstack and product engineers to get work into production
• Read relevant research and test ideas to improve models

Required qualifications

• Currently pursuing a BS, MS, or PhD in Computer Science, Electrical Engineering, or a related field
• Solid foundation in deep learning and computer vision through coursework, research, or projects
• Hands-on experience training models in PyTorch (or similar)
• Strong Python skills
• Curiosity about messy, real-world data: scanned, handwritten, imperfect
• High ownership

Nice to have

• experience with document AI, OCR, or vision-language models

About the company

bobyard is building computer vision applications for the construction trades to automate takeoffs.