Gpu Performance Engineer Jobs in Dallas, TX (NOW HIRING)

Position Summary:

Spectral AI, Inc. (Spectral) is seeking a software engineer to join our growing design and development team. The position is based in Dallas, Texas, with hybrid work flexibility in accordance with the Spectral Employee Handbook guidelines. This position will assist in the integration of computer vision algorithms into our Windows-based DeepView AI software. This role focuses on bridging advanced GPU-based reconstruction and segmentation algorithms with a clinical production software system, ensuring performance, stability, and regulatory readiness.

The software engineer will work closely with Data Science, Algorithm, and UI/UX teams to integrate 2D/3D segmentation, point cloud reconstruction, mesh generation, and visualization into a cohesive, multi-threaded GPU-accelerated workflow.

Key Responsibilities:

3D Reconstruction & Visualization Integration

• Point cloud reconstruction pipelines
• Mesh generation and texturing workflows
• OpenGL-based visualization components

Windows Software & C# Integration

• Integrate GPU-accelerated algorithms into an existing Windows OS–based C# application
• Coordinate C# application logic with native GPU components (CUDA / TensorRT / OpenGL / Open3D)
• Support MVVM-based architecture, async workflows, and UI responsiveness

GPU & Multi-Threaded System Design

• Work with multi-threaded execution models:

• Background CPU threads for capture, processing, and data management
• GPU streams for segmentation, reconstruction, and visualization

• Handle GPU initialization, warm-up, memory allocation, synchronization, and data transfer
• Coordinate CUDA ↔ OpenGL interoperability for real-time visualization

Required Qualifications

• Solid understanding of 3D math (matrices, vectors, quaternions, transformations), Graphics pipelines, GPU architecture
• Hands-on experience with one or more graphics APIs: OpenGL, Direct3D, Vulkan or Metal
• Experience integrating OpenGL/Vulkan with C# via OpenTK, Silk.NET, SharpGL, or custom interop
• Understanding of: OpenGL context creation and lifecycle and WGL / platform-specific context management
• Familiarity C#/.NET/C++, shader languages (HLSL, GLSL) in complex Windows desktop applications
• Experience with multi-threaded programming and async execution models
• Experience with 3D point clouds, meshes, and visualization
• Familiarity with ray tracing or projection of 2D masks into 3D space
• Understanding of mesh texturing, smoothing, and artifact handling is a plus

Preferred Qualifications:

• Experience in medical device software development
• Familiarity with regulated software environments (FDA, IEC 62304, ISO 13485)
• Solid understanding of medical imaging or computational geometry pipelines
• Experience working alongside Data Science / Algorithm teams
• Familiarity with CUDA and Open3D
• Experience handling GPU memory transfer, synchronization, and performance optimization
• Exposure to Qt/C++ or cross-platform UI frameworks (future platform migration is a plus)
• Background in computer vision, photogrammetry, or 3D reconstruction

Required Education and Experience:

• Bachelor’s degree or above in Computer Science, software engineering, Electrical/Computer Engineering, or a related field.
• 3–5 years of proven experience working with 3D graphics, computational geometry, or imaging pipelines involving point clouds, meshes, or segmentation workflows
• Demonstrated proficiency in technical writing, producing accurate, comprehensive, and well-structured documentation for software.

Soft Skills

• Excellent written and verbal communication skills in English.
• Detail-oriented, strong analytical and problem-solving skills.
• Ability to prepare quality software documentation suitable for regulatory review.
• Ability to collaborate across functional teams such as engineering, QA/RA, hardware, and clinical teams.
• Comfortable managing multiple concurrent security tasks.

Physical Requirements:

• Prolonged periods of sitting at a desk and working on a computer.

Off-hours support:

Occasional support during off-hours for security events or release deadlines.