Title
Computer Vision Engineer
Quick Summary
Sightline Robotics is hiring a Computer Vision Engineer to build and productionize perception components that power our next-generation autonomous platforms. You will design and deploy detection, segmentation, multi-object tracking, pose estimation, and 3D perception pipelines, with a strong emphasis on real-time performance and reliability on edge hardware. We welcome strong graduates and early-career engineers who can demonstrate solid projects and a clear grasp of geometry, deep learning, and systems thinking.
Project Category or Industry
Robotics and autonomy for industrial inspection and logistics.
Type
Full-time employment.
Experience Level
Entry to mid-level, with mentorship and clear growth paths. Experienced candidates are also encouraged to apply.
Duration
Permanent role.
Location
Remote-first with optional hybrid collaboration in Austin, TX and Munich, DE. Maintain at least four hours of overlap with teams operating between UTC−6 and UTC+2.
Salary
USD 94,000–138,000 base depending on location and experience, plus benefits and an annual performance bonus.
Payment Mode
Monthly payroll where supported. In select countries, a compliant contractor arrangement can be used.
Hiring Company Name
Sightline Robotics
Required Skills or Tools
Strong Python and proficiency with at least one of C++ or CUDA; practical experience with PyTorch and OpenCV; understanding of multi-view geometry, camera models, and calibration; familiarity with ONNX/TensorRT optimization and deployment on NVIDIA Jetson-class devices; comfort with ROS 2, Docker, and experiment tracking. Experience with data labeling workflows and dataset versioning is a plus.
Project Description
Sightline Robotics builds autonomous systems that perform inspection and perception-driven navigation in dynamic environments. As a Computer Vision Engineer, you will turn product requirements into robust perception services that run on constrained edge hardware. The work spans dataset design and collection, model development, algorithmic optimization, integration with robotic middleware, and careful measurement of quality, latency, and resource use to meet real-world constraints.
Core Responsibilities and Expected Deliverables
Design and implement models for detection, instance and semantic segmentation, keypoint and pose estimation, and 3D depth or occupancy mapping.
Build tracking pipelines that fuse temporal cues for stable identities, re-identification, and trajectory estimation.
Develop stereo and monocular depth modules, camera calibration routines, and coordinate-frame transformations.
Optimize models for edge inference using quantization, pruning, TensorRT engines, and asynchronous pipelines with GStreamer.
Ship production-grade perception nodes and microservices in ROS 2 with clear APIs, tests, and deployment manifests.
Establish offline and online evaluation harnesses with golden datasets, simulation scenarios, and field metrics; deliver dashboards, alerts, and runbooks.
Produce concise technical documentation and partner with robotics, planning, and QA teams during field tests and releases.
Required Experience and Preferred Qualifications
Proficiency in Python, plus C++ for performance-critical components; solid grasp of linear algebra, probability, and estimation.
Hands-on experience with PyTorch (or TensorFlow), OpenCV, and modern detector/segmenter architectures.
Working knowledge of ROS 2, Docker, GitHub Actions, and cloud artifact registries.
Preferred: experience with Jetson Orin/Xavier, TensorRT/ONNX Runtime, camera calibration toolchains, synthetic data generation, and basic sensor fusion with IMU or LiDAR.
Evidence of impact via internships, publications, open-source contributions, or shipped robotics projects will be valued.
Tools or Platforms to Be Used
Modeling and experimentation: PyTorch, torchvision, PyTorch Lightning, scikit-learn, Weights & Biases or MLflow.
Vision and geometry: OpenCV, Kornia, COLMAP tools for calibration and reconstruction as needed.
Edge and middleware: ROS 2 (Humble or newer), NVIDIA Jetson, CUDA, TensorRT, ONNX Runtime, GStreamer.
Services and infrastructure: Docker, GitHub Actions, artifact registries, object storage, and basic cloud services for data ops.
Data ops: Label Studio or CVAT, DVC for dataset versioning, Parquet/Arrow where appropriate.
Language Requirement
Professional English is required. Additional languages are welcome for cross-regional collaboration and field deployments.
Communication Style
Written-first collaboration using GitHub for issues, pull requests, and design reviews; Slack for daily coordination; Zoom for stand-ups, demos, and incident retrospectives. Clear, actionable documentation is expected for all changes.
Time Commitment or Working Window
Standard 40 hours per week with flexible scheduling. Maintain a predictable daily block that overlaps at least four hours with the core team between 09:00 and 17:00 in your local time.
Payment Terms
Salary is paid monthly via payroll. For contractors, invoices are processed on net-30 terms after acceptance of deliverables and timesheets.
Evaluation Criteria
Portfolio or code samples demonstrating real-time perception, geometry, and optimization for edge deployment.
Practical exercise focused on implementing and evaluating a detection or tracking pipeline with performance constraints.
Technical interview covering multi-view geometry, calibration, model optimization, and observability.
Final conversation on product sense, collaboration, and communication.
References may be requested.
Other Requirements
New hires sign a confidentiality agreement and comply with security and data-handling policies. Light time-tracking may be used for distributed coordination. Occasional travel for field tests and customer pilots—typically two to four trips per year—may be required.
About Sightline Robotics
Sightline Robotics is a privately held robotics company building perception-first autonomy for industrial inspection and logistics. Headquartered in Austin with a distributed team across North America and Europe, we combine rigorous engineering with practical field testing to deliver reliable, real-time perception on constrained hardware. Learn more at https://www.sightlinerobotics.com and contact our hiring team at careers@sightlinerobotics.com.