Job Posting – Software Developer, Maps & Navigation
MAZDIS is a Canadian company based in Vancouver, British Columbia with its core specialty and focuses on Secure Automated Bicycle Parking (SABPS). At MAZDIS we combine a variety of sciences and technologies, such as robotics, automation, computing science, AI, and progressive manufacturing and design methods to provide unique, reliable, and efficient integrated bicycle parking solutions. MAZDIS bicycle parking solutions significantly improve the bicycle parking experience and the security of the bicycle and reduces the storage footprint and along with its intelligent software complete the transportation network by integrating personal cycling into other modes of transportation which significantly improves the First and Last Mile problems.
MAZDIS Semi-Autonomous Cargo Trailers enables bicycle as the most economical and environmentally friendly option, to have the towing capacity for a medium to a large load. As an ideal candidate, you are skilled in signal processing, machine learning, optimization, 3-D simulations, computer vision, programming, numerical analysis, image processing, electronics, and communications. You will build systems that process and analyze our diverse semi-autonomous driving data and our neural network training labels to produce analytics. You will research, design, implement, optimize, and deploy deep learning models that advance the state of the art in perception and control for semi-autonomous driving. A typical day to day includes reading deep learning code/papers, implementing described models and algorithms, adapting them to our setting, driving up internal metrics, working with downstream engineers to integrate neural networks to run efficiently on the embedding system, and incrementally tracking and improving feature performance based on the semi-autonomous trailer and leading bicycle.
· Train machine learning and deep learning models on a computing cluster to perform visual recognition tasks, such as segmentation and detection.
· Develop state-of-the-art algorithms in one or all the following areas: deep learning (convolutional neural networks), object detection/classification, tracking, multi-task learning, large-scale distributed training, multi-sensor fusion, etc.
· Optimize deep neural networks and the associated preprocessing/ post-processing code to run efficiently on an embedded device.
· Design the software architecture and firmware implementation on hardware through integration, test, and high-volume manufacturing.
· Build data pipelines and visualizations that provide actionable insights from the data.
· Develop and implement motion planning software and algorithms for autonomous vehicles.
· Model both navigation and obstacle avoidance in a dynamic environment.
· Design and develop autopilot software architecture, including designing interfaces between subsystems
· Collaborate with the control systems, simulation, and modeling teams to design control strategies that can be implemented in software efficiently.
· Conduct high-level system design and analysis
· Design and implement metrics, applications, and tools that provide data insights.
· Master of Computer Science, Computer Engineering, Electrical Engineering with hands-on experience with robotics, real-time controls, system software, or equivalent practical knowledge.
· An ideal candidate is very comfortable in cluster environments and experienced with computer systems concepts (CPU/GPU interactions/transfers, latency/throughput bottlenecks during the training of neural networks, CUDA, pipelining/multiprocessing, etc.).
· We are at the cutting edge of deep learning applications. The ideal candidate has a strong understanding of the under the hood fundamentals of deep learning (layer details, backpropagation, etc), implementing related academic literature and experience in applying state of the art deep learning and reinforcement learning models to visual navigation
· Experience with data science tools including Python scripting, numpy, scipy, matplotlib, scikit-learn, jupyter notebooks, bash scripting, Linux environment.
· Experience with PyTorch, or at least another major deep learning framework such as TensorFlow, MXNet.
· In-depth knowledge of Maps / GIS / Computational Geometry data
· Hands-on experience in robotic or autonomous vehicle system design and implementation: feature development, system debugging and code optimization
· Advanced proficiency with driver assistance sensors such as radar, camera, ultrasonic, and lidar, including the measurement and data-reduction, target identification and environmental synthesis, and sensor fusion.
· Understanding of electrical engineering principles; communication systems, ability to read & understand schematics, use lab instruments, build basic electrical systems, and board /chip bring up.
· Experience with both simulation and hardware implementation.
· Experience with solving software problems using geometry and linear algebra
· Shipped safety-critical software in a fast-pace environment
· Strong understanding of statistics.
· Ability to rapidly prototype and test new algorithms
Impact and desired outcomes
People around the world began using bicycles more often for their commute over the past two decades. It is growing so fast that cycling has become one of the main modes of transportation in most countries. The proposed project is a companion to the storage system line of products, and it arises from the company’s extensive market research and strategic market positioning within the next five years. The company believes that the development of such a trailer would expand and ultimately disrupt the overall cargo transportation market. The new trailer is envisioned to address the needs of a larger population. It will also encourage resource sharing via the cloud with other users. Environmentally, the new trailer will be an electric vehicle, and thus, it is more economical and environmentally friendly than the conventional towing truck.
Interaction During COVID-19
You can work from HOME, as long as:
Attend Daily the 5-minute Meetings (call during COVID-19)
Attending Product Development Meetings (Video Conference during COVID-19)
Be fully Engaged with Supervisor as required
We would like to receive progressive reports during your engagement. Your reports will be incorporated into the guidelines and procedures of the company.
Please send your application (cover letter, resume and portfolio) to HR@mazdis.com.