Senior ML Computer Vision Engineer

Plenoptic Cognition Inc.; United States; Remote; USD 180.000 - 260.000; Vollzeit

Benefits dieser Stelle

Competitive base salary + stockOptions

Health, dental, vision, and 401(k)

Flexible remote work with occasional,

Zusammenfassung

VectorField is seeking a senior ML vision specialist to design, implement, and scale core perception models powering high-fidelity 2-D to 4-D visual pipelines.

You will work on CNNs, transformers, diffusion, NeRF, and production-grade services on AWS GPUs, while ensuring security and provenance per CMMC/ITAR. Join a team that blends research with robust deployments.

Qualifikationen

• MS or PhD in CS/EE/Applied Math with focus on computer vision/graphics.
• 7+ years of professional ML/vision work, including at least 3 years leading end-to-end production systems.
• Deep expertise in digital-avatar pipelines, deep-fake detection, 3-D reconstruction, SfM, multi-view stereo.
• Proficiency in Python, PyTorch/TensorFlow, and C++/CUDA for custom kernels.
• Experience with ONNX/TensorRT, Triton Inference Server, or equivalent deployment stacks.
• Familiarity with container orchestration (ECS/Fargate or Kubernetes).
• Understanding of Hyperledger Fabric or other permissioned ledger tech for immutable data logs.

Aufgaben

• Lead design of large-scale vision models (CNN, Transformer, Diffusion, NeRF) for 2-D → 3-D → 4-D conversion and publish internal white-papers.
• Define pipelines ingesting multi-modal sensor data (RGB, depth, LiDAR, hyperspectral, neuromorphic) and produce clean labeled datasets.
• Translate research prototypes into production-grade services on AWS GPU clusters using Go/Python micro-services, TensorRT/ONNX runtime.
• Collaborate with Digital Twin/ABM teams to feed model outputs into feedback loops.
• Ensure training data complies with security and classification markings (CMMC, ITAR).
• Mentor junior vision engineers and data scientists, establishing coding standards and >80% test coverage.

Kenntnisse

Python

PyTorch

TensorFlow

C++/CUDA

Security clearance

AWS

Kubernetes

Ausbildung

MS/PhD in CS/EE/Applied Math

Tools

ONNX/TensorRT

Triton Inference Server

Docker/Kubernetes

SageMaker

Jobbeschreibung

VectorField's Autopoietic Tensor-as-a-Service (aTaaS) platform hinges on high-fidelity visual data pipelines that turn raw sensor streams into reusable AI training assets (digital avatars, deep-fake detectors, 3-D/4-D reconstructions, etc.).

We need a senior ML vision specialist who can design, implement, and scale the core perception models that power these pipelines while respecting the strict security and provenance requirements of our platform.

Key Responsibilities

• Model Architecture & Research: Lead the design of large-scale vision models (CNN, Transformer, Diffusion, NeRF, etc.) for 2-D → 3-D → 4-D conversion. Publish internal white-papers and external research
• Data Engineering & Curation: Define pipelines that ingest multi-modal sensor data (RGB, depth, LiDAR, hyperspectral, neuromorphic) and produce clean, labeled datasets
• Algorithmic Production: Translate research prototypes into production-grade services (Go/Python micro-services, TensorRT/ONNX runtime) on AWS GPU clusters
• Cross-Domain Integration: Collaborate with the Digital Twin / ABM team to feed model outputs into the Bits2Effect feedback loop
• Security & Compliance: Follow operational-security best practices; ensure all model training data complies with CMMC, ITAR, and applicable classification markings
• Mentorship & Leadership: Guide junior vision engineers and data scientists, establishing coding standards and test coverage (> 80%)

Required Qualifications

• Citizenship: U.S. citizen (must be able to obtain/maintain a Secret clearance)
• Education: M.S. or Ph.D. in Computer Science, Electrical Engineering, Applied Mathematics, or related field with focus on computer vision/graphics
• Experience: 7+ years of professional ML/vision work, including at least 3 years leading end-to-end production systems
• Deep expertise in digital-avatar pipelines, deep-fake detection, 3-D reconstruction, Structure-from-Motion (SfM), multi-view stereo
• Proficiency in Python, PyTorch/TensorFlow, and C++/CUDA for custom kernels
• Experience with ONNX/TensorRT, Triton Inference Server, or equivalent deployment stacks
• Familiarity with container orchestration (ECS/Fargate or Kubernetes)
• Understanding of Hyperledger Fabric or other permissioned ledger tech for immutable data logs

Preferred Qualifications

• Publications in top CV conferences (CVPR, ICCV, ECCV) on avatar synthesis, deep-fake mitigation, or 3-D reconstruction
• Hands-on experience with Neural Radiance Fields (NeRF), Gaussian Splatting, or Diffusion-based 3-D generation
• Prior work integrating vision models with digital twins or agent-based simulations
• Experience with large-scale data pipelines on AWS (S3, SageMaker, EMR)
• Knowledge of formal verification (TLA+, model checking) as applied to ML inference safety

What We Offer

• Competitive base salary + performance-based equity (stock options)
• Comprehensive health, dental, vision, and 401(k) matching
• Flexible remote work policy with occasional secure-site visits (travel expenses covered)
• Access to cutting-edge GPU clusters, proprietary datasets, and a collaborative research environment
• Opportunity to shape a platform that could become the backbone of future AI-driven visual media

Shape the core perception models that power the next generation of AI-driven visual media and defense-grade perception systems.

Learn more about the VectorField Platform