About the role
Data Scientist – Python & Computer Vision
Location: McLean, VA – Fully Onsite
Duration: Contract through December 31, 2026
Assignment Type: Contract Only
Work Authorization: Sponsored candidates accepted through one-layer-deep sub-vendors; candidates must be W2 employees of the sub-vendor.
Interview: 1 round, 60 minutes – MS Teams video mandatory
Industry: Mortgage / Financial Services
Important Submission Requirement
Completed resume templates and candidate vetting questions are required with every submission. Resumes without completed templates and/or vetting questions will not be considered. All vetting-question responses must be provided directly by the candidate.
About the Role
We are seeking a hands-on Data Scientist with strong Python, Computer Vision, SQL, data modeling, and Snowflake experience to develop and operationalize image-based data capabilities within a mortgage/financial-services environment.
The role will focus on image extraction, computer vision model outputs, metadata generation, validation, quality control, and structured data integration . The ideal candidate combines strong software engineering and data skills with experience working with image analytics and mortgage/financial data.
Key Responsibilities
- Develop Python solutions supporting computer vision, image extraction, metadata generation, and model-output validation .
- Process image-based model outputs including bounding boxes, labels, confidence scores, attributes, and structured metadata .
- Build reusable Python utilities for parsing, transforming, validating, and comparing computer vision outputs.
- Create test datasets and perform regression testing and quality validation across model runs.
- Validate extracted images and metadata against PDFs, appraisal reports, and authoritative source documents .
- Identify defects, data gaps, extraction issues, misclassifications, and model-output inconsistencies.
- Develop SQL/Python pipelines to transform model outputs into structured and semi-structured data.
- Work with JSON/Variant data structures and Snowflake for downstream analytics.
- Partner with product, engineering, UI, research, modeling, data, and business teams.
- Translate technical model outputs into practical QC, review, validation, and exception workflows .
- Troubleshoot issues across image inputs, model outputs, metadata, data loads, and user-facing workflows.
- Document technical findings, defects, remediation steps, and testing results.
Required Qualifications
- 5–7+ years of related experience in Data Science, Computer Vision, Data Engineering, or a similar technical discipline.
- Strong hands-on Python development experience.
- Strong SQL skills and experience working with structured and semi-structured data.
- Hands-on Computer Vision / Image Analytics experience.
- Experience with libraries such as OpenCV, Pillow, PyTorch, TensorFlow , or similar.
- Experience working with image files, model predictions, labels, confidence scores, bounding boxes, annotations, and metadata .
- Strong experience with data modeling and Snowflake .
- Experience with JSON, APIs, data pipelines, and validation/reconciliation logic.
- Experience testing, debugging, and validating model-output pipelines.
- Ability to work with business and technical stakeholders in a fast-paced environment.
- Financial services or mortgage industry experience is required.
Preferred Qualifications
- Experience with Freddie Mac or Fannie Mae .
- Mortgage, appraisal, property-image, or housing data experience.
- Experience with OCR, object detection, image classification, image extraction, or metadata extraction .
- Snowflake VARIANT/JSON experience.
- Experience with model validation and AI-enabled quality-control workflows.
- Knowledge of UAD/appraisal modernization .
- Experience working with GSE, mortgage, housing, or financial-services technology.
⭐ Top 5 Must-Have Skills
- Python – Strong hands-on development
- Computer Vision / Image Analytics
- SQL + Data Modeling
- Snowflake + JSON/Semi-Structured Data
- Mortgage / Financial Services Experience
Millions of jobs, with real people getting hired every day
Questions, answered
Click "Apply with JobAssist" – we tailor your resume and application to this role and submit it for your approval.
Yes. This role at PTR Global was screened before publishing – we confirmed the employer before listing it.
The employer didn't disclose a salary range for this listing. JobAssist shows pay whenever it's available.
This position can be done from anywhere, with no in-office requirement.
Yes – every application is tailored from your profile and this job's requirements, and you can review and edit before it's sent.