Back to results

Senior Embedded DSP & Machine Learning Engineer

  • Montreal, QC, Canada

Full-time

Apply on LinkedIn About employer

Applications open in a new tab.

Job details

Job type
Full-time
Apply by
Experience
At least 5 years of professional experience in embedded software development, preferably involving one or more of: real-time, signal-processing, compute-constrained systems
Posting language
English
Posted

Original job description

Company Overview

Faurecia IRYStec is a highly innovative technology company, maintaining a startup feel within a big company. Our goal and focus are to revolutionize user experience in the cabin through the improvement of cabin monitoring systems and automotive displays.

We have a multi-disciplinary team of researchers and developers working closely with Academia and Industry. We cooperate with a global team of professors in well-known universities and recognized industry leaders. We are part of the FORVIA group formed following the successful completion of the acquisition of a controlling stake in HELLA by Faurecia on January 31, 2022. This new Group represents 150,000 talented individuals across more than 40 countries.

The Faurecia IRYStec team is composed of software & hardware engineers, as well as signal processing and AI researchers, who work collaboratively. Our software solutions encompass multiple platforms such as GPU, embedded systems, PC, and cloud architectures. We are also engaged in implementing our solutions into FPGA and ASIC devices.

Innovation is part of our DNA; we demonstrate every year new Proof of Concepts at major industry conferences, tradeshows and internal events with top executives' audience.

Job Description

We are looking for a Senior Embedded DSP & Machine Learning Engineer to accelerate the development of our in-cabin sensing platforms.

The ideal candidate has a strong background in both machine learning and embedded software engineering, having brought to production ML-based solutions into DSP equipped solutions, including by starting from trained models or algorithms from higher level languages (e.g. Python).

The primary responsibilities of the Embedded DSP & Machine Learning Engineer are to identify bottlenecks and optimize the performance of the overall processing pipeline focusing on key metrics such as latency, memory and code size, computing resources utilization and fitness for production.

This role involves close collaboration with ML researchers, system engineers, and perception teams, and focuses on advancing performance of in-cabin sensing systems based on radars and cameras.

This position reports to the Director of Engineering.

What you will do:

  • Implement DSP pipelines and port higher level software algorithms as invented by our team.
  • Optimize ML algorithms to make the best use of the target embedded platform in terms of resources and performance
  • Debug complex issues across software, algorithms, machine learning models and hardware platforms
  • Port and optimize algorithms and machine learning inference pipelines to diverse heterogeneous platforms
  • Evaluate algorithm performance against system-level requirements (e.g., detection accuracy, false alarm rate, latency)
  • Collaborate with ML teams to optimize models for size, including tradeoffs measurement and quantification
  • Participate in scrum and code reviews
  • Contribute to technical documentation, publications, and patents

Requirements:

  • Bachelor’s or Master’s degree in Computer Engineering, Electrical Engineering, Computer Science, or a related technical field
  • At least 5 years of professional experience in embedded software development, preferably involving one or more of: real-time, signal-processing, compute-constrained systems
  • Strong expertise in C programming, especially low-level development (memory, pointers, data structures) and performance-critical code
  • Experience developing software for embedded processors, MCUs, DSPs, SoCs, and heterogeneous computing platforms
  • Strong understanding of embedded architectures along following keywords: memory hierarchy, multicore, fixed and floating-point, SIMD, real-time
  • Strong understanding of machine learning and deep learning including model architectures, training, validation and inference
  • Hands-on experience deploying and executing trained ML models on embedded edge platforms in a production environment
  • Experience converting models from higher level frameworks (PyTorch, TensorFlow) to optimized runtimes
  • Experience with such technologies as ONNX, TensorFlow Lite, TensorRT, or equivalent embedded inference frameworks
  • Experience working with RTOS and embedded Linux environments
  • Strong understanding of interfaces, communication, memory management, synchronization and real-time execution

Pluses:

  • Proven hands-on experience developing, profiling and optimizing DSP code not only for traditional metrics but also power and processing efficiency
  • Experience with common DSP techniques such as FFT, filters, matrix operations, signal conditioning, feature extraction and DSP pipelines
  • Ability to analyze and optimize ML inference performance across architectures (CPU, DSP, GPU, NPU) including techniques such as distillation, quantization, pruning, low-level optimization, hardware acceleration
  • Ability to benchmark different models and make tradeoffs between accuracy, latency and resources usage
  • Experience with automotive or related transportation systems
  • Experience with Qualcomm, TI, NXP, NVIDIA, or similar ARM platforms
  • Knowledge of NEON, SIMD, CUDA, OpenCL or other hardware specific acceleration techniques
  • Familiarity with embedded software development practices including Git, CI/CD, unit testing, static code analysis, and automated verification

We provide:

  • Hybrid Work
  • Flex Vacation: We offer a flexible vacation policy, encouraging employees to take four weeks of vacation annually to support their well-being and work-life balance. Additional time off may be approved by the GM, subject to project deadlines and business needs. 
  • Generous Health Benefits & Pension Plan 
  • Learning environment
  • In-house specialists in different fields are ready to share their knowledge.
  • Training, conferences, and books. We want you to grow with us.
  • Diverse environment
  • We put a great deal of effort into fostering a diverse and inclusive workplace that values and embraces individuals of all genders, races, ethnicities, backgrounds, and experiences.
  • You can hear our Founder talk about the technology and diversity here (https://www.youtube.com/watch?v=goIs57t9dv4&t=88s)
  • Great environment: Great teamwork, great energy, and a workplace where we genuinely enjoy being together.
More about this job

Responsibilities

Implement and optimize DSP pipelines and machine-learning inference for embedded platforms, improving latency, memory use, code size, and computing efficiency. Debug issues across software, algorithms, models, and hardware; evaluate system-level performance; and collaborate with research and engineering teams on production-ready in-cabin sensing systems.

Requirements

Requires a bachelor’s or master’s degree in a relevant technical field and at least five years of embedded software development experience. Candidates should have strong C and embedded architecture skills, practical experience deploying and optimizing machine-learning models on edge platforms, and familiarity with embedded inference frameworks, RTOS, and embedded Linux.

Skills

  • C Programming
  • Embedded Software Development
  • Digital Signal Processing
  • Machine Learning
  • Deep Learning
  • Model Optimization
  • Embedded Machine Learning Inference
  • Real-Time Systems
  • Embedded Linux
  • RTOS
  • Performance Profiling
  • Memory Management
  • SIMD
  • Fixed- And Floating-Point Computing
  • ONNX
  • TensorFlow Lite

Benefits

  • Hybrid Work
  • Flexible Vacation
  • Health Benefits
  • Pension Plan
  • Training And Development
  • Conferences
  • Books

Visa sponsorship

Not detected in the job text

Categories

  • Engineering
  • Software
  • Technology
  • Science & Research

Keywords

  • C
  • Python
  • PyTorch
  • TensorFlow
  • ONNX
  • TensorFlow Lite
  • TensorRT
  • DSP
  • Machine Learning
  • Deep Learning
  • Embedded Systems
  • Embedded Linux
  • RTOS
  • MCUs
  • SoCs
  • Heterogeneous Computing
  • Memory Hierarchy
  • Multicore Processing
  • Fixed-Point Computing
  • Floating-Point Computing
  • SIMD
  • Real-Time Systems
  • Model Quantization
  • Model Pruning
  • Knowledge Distillation
  • FFT
  • Filters
  • Matrix Operations
  • Automotive Sensing
  • Radar And Camera Systems

Report this job

Get in touch
MovoJobs

How can we help?