Senior Computer Vision Engineer (Scientific Imaging)

Location: Remote, Tbilisi

Company

Develops projects for clients including PayPal, Wargaming, Xerox, Philips, adidas, and Toyota.

What you will do

• Lead the two-person computer-vision team developing an X-ray inspection proof of concept for wind-turbine blades.
• Assess initial images, determine visibility and contrast, and define image-quality requirements for the acquisition partner.
• Build preprocessing for tiling, thickness normalisation, denoising, stitching-seam handling, and full-image coordinate reconstruction.
• Develop conductor tracing using line and ridge detection, graph-based tracing, branching logic, and rejection of look-alike structures.
• Run feasibility experiments for detecting missing adhesive or open separations using simulators and reference samples with known defects.
• Own validation, thresholds, uncertain-case policy, technical decisions on misses, implementation reviews, and technical sign-off.

Requirements

• 4+ years of experience in computer vision or image processing on real, imperfect images.
• Working knowledge of classical image processing, including filtering, ridge and line detection, morphology, segmentation, connected components, and graph-based tracing.
• Experience processing images too large for memory, including tiling, overlap, and coordinate reconstruction.
• Strong Python with NumPy, OpenCV, scikit-image, or equivalent tools.
• Experience designing labelling and validation methods for scarce data, quantifying image quality, and separating imaging limitations from algorithm limitations.
• English sufficient for reading technical papers and communicating with client engineers.

Nice to have

• Experience with radiography, CT, NDT, composites, weld inspection, or semiconductor inspection.
• Physics or engineering background, imaging simulation experience, or experience with small-dataset deep learning.
• Experience influencing physical data acquisition and using AI coding agents for prototyping and tooling.

Culture & Benefits

• Six- to eight-month proof of concept with a path to a product.
• Flexible working hours aligned with the employee's schedule and remote-work options.
• Medical-expense compensation and five paid sick days per year without requiring a sick-leave certificate.
• Online English courses, career development with clear skill qualifications, and gym-membership compensation.
• Corporate parties, events for employees and their children, and corporate sports competitions.
• One approximately one-week on-site test campaign at a wind farm is required.