About the role
The CV/ML Engineer will help build and improve AI systems for large-scale video and vision workloads in a real-world security setting. You will work with senior engineers to develop, train, and deploy computer vision models that run across edge and cloud environments.
You’ll contribute to model development, infrastructure, and tooling that support reliable, production-grade ML services used in safety-critical scenarios. This role is a good fit for someone who has hands-on CV/ML experience and wants to grow into owning larger parts of the stack over time.
Responsibilities
Implement and maintain computer vision models for detection, tracking, and related tasks on video streams.
Support end-to-end ML workflows: data preparation, training, evaluation, and deployment.
Help integrate models into low-latency services and APIs used in real-time security products.
Assist in monitoring model performance and improving robustness using real-world data.
Contribute to basic optimization work such as model compression, quantization, and efficient runtime configuration.
Collaborate with senior engineers on infrastructure, security, and networking improvements for deployed ML systems.
Qualifications
2 to 4 years of experience in computer vision or machine learning engineering (industry or strong research/production projects).
Practical experience training and evaluating deep learning models in PyTorch or TensorFlow.
Some exposure to running models in production (scripts, simple services, or pipelines) and working with cloud infrastructure (AWS or similar).
Familiarity with Docker and basic concepts of containerized deployments; Kubernetes experience is a plus.
Understanding of the data science lifecycle from dataset creation to deployed model monitoring.
Basic understanding of networking and security concepts (e.g., APIs, VPCs, IAM, authentication).
Comfortable working in a small, fast-moving team, receiving feedback, and iterating quickly.
Bonus skills
Experience with video pipelines or tools such as GStreamer, DeepStream, OpenVINO, or FFmpeg.
Knowledge of GPU acceleration, edge devices, or IoT-style deployments.
Exposure to MLOps practices (CI/CD for ML, experiment tracking, or model registries).
Why join
Work on real-world computer vision systems that directly impact physical security and incident prevention at scale.
Learn from and collaborate closely with experienced engineers on distributed inference, infrastructure, and product.
Join an ambitious, venture-backed security startup early and grow your responsibilities as the company scales.
Compensation
An equity package aligned with early founding team members, with standard vesting terms.
