Job VC

Crunch-IS is looking for an experienced
Computer Vision / Machine Learning Specialist
to support the development of advanced perception systems for autonomous platforms.
You will contribute to the design, evaluation, and optimization of real-time computer vision and machine learning algorithms that operate on edge hardware and process real-world sensor data. The role involves applied research, rapid experimentation, and practical implementation of robust AI solutions capable of performing reliably in dynamic environments.
This position is ideal for professionals who combine strong theoretical foundations with hands-on engineering experience and can provide practical technical guidance across the full lifecycle of CV/ML solutions.
Responsibilities:
Provide expert guidance on the design and implementation of
computer vision and machine learning models
.
Develop and optimize algorithms for
object detection, tracking, classification, and scene understanding
.
Design
real-time perception pipelines
for processing video and multimodal sensor data.
Support deployment of ML models on
embedded and edge hardware platforms
.
Evaluate model performance and recommend improvements in
accuracy, robustness, latency, and efficiency
.
Define approaches to
dataset preparation, annotation pipelines, and validation methodology
.
Optimize models for operation in
resource-constrained environments
.
Collaborate with cross-functional engineering teams to ensure practical and scalable solutions.
Provide technical recommendations on architecture, frameworks, and development workflows.
Contribute to improving the reliability of perception systems operating in complex real-world conditions.
Support transition of solutions from
R&D prototypes to production-ready systems
.
Requirements:
Strong expertise in
Computer Vision and Machine Learning
.
Proven experience developing and deploying
real-world CV/ML solutions
.
Proficiency in:
Python and/or C++
OpenCV or similar computer vision libraries
PyTorch, TensorFlow, or similar ML frameworks
Experience with
embedded AI platforms
(e.g., Jetson, ARM-based devices, NPUs).
Experience working with
real-time video or multimodal sensor data
.
Experience with
multimodal perception
(e.g., video, telemetry, acoustic, or other sensor signals).
Knowledge of
3D perception, SLAM, or sensor fusion techniques
.
Experience optimizing models for
edge deployment
under compute and latency constraints.
Experience integrating perception models into
autonomous or semi-autonomous systems
.
Practical experience bringing
R&D prototypes into production-ready solutions
.
Understanding of trade-offs between
accuracy, latency, and hardware limitations
.
Strong analytical thinking and ability to solve complex technical problems independently.
Nice to have:
Experience with UAV autopilot stacks such as
PX4 or ArduPilot
.
Experience working in
aerospace, robotics, or deep-tech environments
.
Experience designing
end-to-end perception pipelines
.
Experience with model optimization techniques (quantization, pruning, TensorRT, etc.).