Job ID 21338
Sector Machine Learning
- On-site parking
- Referral programme
Machine Learning Scientist – Part-time research contract
At Archangel Imaging, you will be working alongside a fun, experienced and forward-thinking team to deliver transformative AI solutions that help organisations like Network Rail, Nuclear Decommissioning Authority and Ministry of Defence better protect our service personnel, critical infrastructure and first responders over large, remote areas. But that’s not all! We work extensively with industry drones and robotics partners to create turn-key solutions, so you will have exposure to the latest cutting-edge technologies. You will be joining a company that is enjoying great success. We have recently won the GENIUS NY program in New York (the largest accelerator program for unmanned systems in the world). In addition, we have won several exciting development projects with key defence customers.
Joining us as a Machine Learning Scientist, you will have an opportunity to push the boundaries of autonomous drone navigation in a friendly, cutting-edge technology company. Following a successful research project in autonomous drone navigation we are looking to expand our team with the aim to bring our solution as a product to market. We are also continuing our research into visual navigation to keep improving our product and need talented engineers and scientists to help drive our world-leading research.
Our drone platform will operate autonomously in challenging environments where no GPS or other navigation signals are available, in civil and military use cases. Our team develops hardware and software to allow drones to navigate using visual information (such as visual odometry, image registration, and terrain recognition). Furthermore, just like humans, drones need also need to be able to navigate at night. Therefore, we are looking for a Machine Learning Scientist who wants to tackle the challenges of night-time visual navigation using state-of-the-art deep learning algorithms.
What you’ll be owning:
- Researching and developing state-of-the-art algorithms to assist visual navigation at night, with a particular focus on thermal-to-thermal and thermal-to-rgb registration
- Working on solutions that can be deployed both to the cloud (no computational constraints) and in embedded systems (strict computational constraints)
- Testing your algorithms on flight data and analysing results to inform future improvements
- Keeping the Algorithms & Simulations Team updated on the latest scientific literature and conferences on visual navigation and deep learning.
We rely heavily on 3D simulations to gather training data and test our solutions so you will be expected to handle large volumes of data and leverage the power of cloud computing to do so. For the development of our deep learning models, we use PyTorch and the Python ecosystem, and for the deployment and testing of our solution we use a combination of Python / C++, ROS, Docker, and Gazebo simulations. As a candidate you will ideally already have experience working with some of these tools but, most importantly, you should be eager to learn. As part of your job, you will be actively trained to be able to work with all these tools.
To succeed in this role, you must have:
- Master’s degree / PhD in deep learning for computer vision
- Excellent Python coding skills
- Experience with at least one of the major Python deep learning frameworks (PyTorch, Tensorflow, Keras) with a strong portfolio of development examples
- Excellent skills in deep learning (CNN training, testing, evaluation and deployment)
- Researching, designing, and implementing machine learning and computer vision solutions
Bonus if you have:
- Experience with thermal and/or night-time computer vision algorithms
- Experience with a variety of computer vision tasks like object recognition, object detection, semantic segmentation
- Experience developing AI / ML solutions for robotics platforms using ROS
- Navigation and camera pose estimation algorithms: image registration, visual odometry, stereo odometry, SLAM, etc.
- Experience deploying deep learning models to embedded and edge hardware (NVIDIA Jetson / mobile platforms)
- Familiarity with developing scalable machine learning pipelines in the cloud (AWS / GCP / Azure)
- Solid understanding of classical computer vision techniques (OpenCV, SVM, KNN, etc.)
- Experience with robotics simulation platforms for data gathering (Gazebo, Airsim, Unity, etc.)
- Software development using C, C++, C#
- Experience working with sensors (cameras, IMU, etc.) and how to get the most out of them
- Eligibility and willingness to undergo future full security clearance (SC)
Working with us means you will have:
- The ability to make a measurable difference in a small company building cutting edge technology with big vision
- Fast-paced environment with a positive, talented team
- Forward-thinking, supportive culture with Monday paid lunch, quarterly company retreats and strategic alignment, flexible working hours and custom arrangements that matter to you
- Exciting growth opportunities and training resources
- Competitive compensation
Hybrid. For this part-time role, you will be expected to work in the office once a week. We offer flexible work arrangements and even fly international team members over to join us on-site for special occasions and all-company get-together
So, what’s next?
Apply now and someone from our HR team will get back to you asap. The usual process includes a screening call, offline assessment and then a big call with our team to get to know each other. That’s it!
Sounds interesting? Apply now and join Archangel Life!
Job Types: Part-time, Temporary contract
Contract length: 6 months
- On-site parking
- Referral programme
- Monday to Friday
Work Location: In person
Expected start date: 01/01/2024