Software Developer – Python, Computer Vision & .NET Backend

About the Role

We are looking for a Software Developer with strong Python skills, practical experience in computer vision or document processing, and solid C#/.NET backend development capabilities.

You will work on a platform that analyzes HVAC and engineering drawings, extracts symbols and technical information from PDFs, and transforms complex documents into structured data that users can review and use.

A significant part of this role involves developing and improving the core extraction algorithms. You will work with PDF text and vector geometry, optical character recognition (OCR), image processing, symbol matching, and AI vision services to improve accuracy across different drawing formats and layouts.

You will also contribute to the .NET backend, including APIs, databases, document management, asynchronous processing, and integration with the Python extraction services.

We are looking for someone who is hands-on, analytical, collaborative, and comfortable investigating difficult technical problems and taking improvements through implementation and validation.

This position follows a hybrid work model, with 2 days per week in our Boucherville office and the flexibility to work remotely for the remainder of the week.

What You Will Do

- Develop, maintain, and improve Python algorithms for extracting information from technical drawings and PDFs.
- Improve legend detection, symbol extraction, symbol matching, and association of symbols with descriptions, equipment tags, dimensions, and airflow information.
- Combine native PDF text and vector geometry with OCR, image processing, and AI-assisted analysis.
- Investigate missed detections, incorrect matches, duplicate results, and ambiguous annotations.
- Develop approaches that work across varied drawings, layouts, scales, rotations, and document quality.
- Build evaluation datasets and regression tests to measure extraction accuracy and prevent regressions.
- Optimize processing speed, memory usage, and external AI service costs.
- Design and maintain backend services and REST APIs using C# and .NET.
- Work with SQL databases, Entity Framework Core, data modelling, and query optimization.
- Improve background processing workflows, message queues, service integrations, retries, and error handling.
- Write clean, maintainable, testable code and participate in code reviews.
- Collaborate with technical teams and domain specialists to translate engineering needs into practical software solutions.
- Document algorithm behavior, technical decisions, and known limitations.

What We Are Looking For

Core Requirements

- Typically 3–5 years of relevant software development experience, or equivalent demonstrated experience delivering substantial software projects.




- Strong professional experience with Python.
- Practical experience with computer vision, image processing, OCR, or document extraction.
- Experience using OpenCV, NumPy, or comparable libraries.
- Working experience developing backend services and REST APIs with C# and .NET.
- Experience with relational databases such as SQL Server.
- Good understanding of algorithms, data structures, and geometric or spatial reasoning.
- Ability to work with coordinates, bounding boxes, image transformations, matching scores, and confidence thresholds.
- Experience with automated testing, debugging, performance profiling, and Git.
- Strong analytical skills and the ability to investigate problems systematically.
- Ability to take a requirement from initial discussion through implementation and validation.
- Clear communication with technical and non-technical stakeholders.

We welcome candidates whose strongest experience is in Python and extraction algorithms, provided they can also contribute to the .NET backend. AI & Document Processing Experience

We are especially interested in developers who have practical experience with one or more of the following:

- Extracting text, tables, symbols, or structured information from PDFs.
- Processing engineering drawings, schematics, or CAD-generated documents.
- OCR integration and reconciliation of recognized text with native document content.
- Template matching, image segmentation, connected components, or shape comparison.
- Associating detected objects with nearby text and annotations.
- Multimodal AI and vision model integration.
- Structured AI outputs and validation.
- Measuring precision, recall, false positives, and false negatives.
- Combining deterministic algorithms with AI services to improve reliability.

You should be comfortable implementing and evaluating extraction logic as well as integrating external AI services.

Technical Skills

Extraction & Algorithms

- Python
- OpenCV and NumPy
- PDF processing libraries such as PyMuPDF
- OCR and document layout analysis
- Image processing and geometric reasoning
- Symbol detection and matching
- Algorithm evaluation and performance optimization

Backend

- C# and .NET
- ASP.NET Core and REST APIs
- Entity Framework Core
- Authentication and authorization
- Background processing and asynchronous workflows
- Service integration and error handling

Databases & Infrastructure





- SQL Server or equivalent relational databases
- Data modelling and query optimization
- Object storage
- Message queues and event-driven processing
- Docker and containerized services

Nice to Have Experience with any of the following would be considered an asset:

- Azure AI Document Intelligence
- Azure AI Foundry or comparable AI platforms
- Azure Blob Storage and Azure Service Bus
- RabbitMQ / Service Bus
- Clean Architecture and separation of application, domain, and infrastructure concerns
- Processing large or complex technical PDFs
- HVAC, mechanical engineering, or construction terminology
- Angular familiarity for collaboration with the frontend team

Education

- Bachelor’s degree in Computer Science, Software Engineering, Computer Engineering, or a related discipline.

OR
- Equivalent professional experience demonstrating strong software development and algorithmic problem-solving capabilities.

What Makes Someone Successful in This Role? The ideal candidate is someone who:

- Enjoys investigating why an extraction algorithm succeeds on one drawing and fails on another.
- Uses measurable results to evaluate improvements.
- Can balance accuracy, processing time, resource consumption, and maintainability.
- Understands when to use deterministic logic, computer vision, or AI services.
- Is comfortable moving between Python algorithms and .NET backend development.
- Considers failure cases and handles uncertain or incomplete results thoughtfully.
- Takes ownership and follows problems through to resolution.
- Communicates clearly and values collaboration, code quality, and knowledge sharing.

Why Join Us?

- Work on software that solves practical engineering and document-processing problems.
- Directly influence the core extraction algorithms and backend architecture.
- Build with a modern stack spanning Python, .NET, computer vision, cloud services, and applied AI.
- Participate in technical decisions and propose improvements.
- See the impact of your work through measurable improvements in accuracy, speed, and user experience.
- Collaborate with a team that values technical curiosity and strong engineering practices.
- Continue developing your expertise in document intelligence, algorithms, and backend systems.
- Benefit from a hybrid work model with 2 days per week in the office.

Benefits

- Casual dress
- Dental care
- Disability insurance
- Extended health care
- Life insurance
- On-site parking
- Paid time off
- Vision care

Work Location: Hybrid remote in Boucherville, QC J4B 1E7 Pay: $90,000.00-$110,000.00 per year

Benefits

- Casual dress
- Dental care
- Disability insurance
- Extended health care
- Life insurance
- In office parking
- Paid time off
- Vision care

Work Location: Hybrid remote in Boucherville, QC J4B 1E7

📌 Software Developer – Python, Computer Vision & .NET Backend (Boucherville)
🏢 Qarrot
📍 Boucherville