14 days ago
Remote, France or Paris, FranceSenior
Responsibilities
- Profile and analyze machine learning models to identify computational, memory, and data-movement bottlenecks.
- Evaluate trade-offs between model quality and computational cost, including latency, throughput, and memory footprint.
- Develop hardware-aware model optimization and neural network architecture search methodologies using real hardware measurements.
- Apply quantization, mixed-precision inference, distillation, and other model compression techniques, including analysis of numerical differences.
- Develop and maintain model export, benchmarking, and deployment pipelines across PyTorch, ONNX, TensorRT, and similar inference runtimes.
- Evaluate deployment strategies and map models onto CPUs, GPUs, and dedicated AI accelerators.
- Work across machine learning research, embedded software, and product engineering teams.
Requirements
- Strong understanding of machine learning and deep neural network architectures with hands-on experience developing models using frameworks such as PyTorch.
- Proficiency programming in Python.
- Experience analyzing neural network computational characteristics and the impact of model architecture on inference performance.
- Familiarity with model architecture search, model scaling, quantization, mixed-precision inference, knowledge distillation, or other model compression methods.
- Experience with machine learning inference and deployment technologies such as ONNX, TensorRT, or similar frameworks.
- Ability to reason across model architectures, computational graphs, inference runtimes, and hardware execution.
- Familiarity with Git, unit testing, debugging, and profiling.
- Strong communication and cross-functional collaboration skills.
- Fluency in written and spoken English and French.
- Modern C++ proficiency is a bonus.
- Familiarity with CUDA or OpenCL is a bonus.
- Experience deploying machine learning models in embedded systems is a bonus.
- Experience mentoring team members on software development and best practices is a bonus.
Benefits
- Competitive base salary of €75,000 to €90,000 annually, depending on experience, plus equity.
- Hybrid work model with three office days per week on Monday, Tuesday, and Thursday, and remote work for the remainder.
- Modern office in central Paris.
- Partial reimbursement for public transportation where applicable.
- Subsidized meal vouchers.
- Wellness Pass benefit.
About Zendar
Zendar builds high-resolution radar and AI-enabled perception software for autonomous systems, enabling accurate object detection and mapping in challenging conditions. It offers a radar-centric autonomy stack, sensor-fusion algorithms, and an evaluation kit/SDK used by vehicle and robotics developers to integrate and evaluate its technology. Privately held and founded in 2017, the company is headquartered in Berkeley, California.