iSIZE specializes in deep learning for video delivery and has developed a deep perceptual ‘precoder’ – a software solution that uses AI-trained to optimize visual quality in order to save video bitrate during encoding. Its flagship product, BitSave, available both as a SaaS platform at bitsave.tech and for on-premise use, reduces encoding bitrates by up to 40% without compromising perceptual video quality. This results in substantial data, energy and cost savings for both the company hosting the video and the end consumer. You can see a presentation about our work in the AOMedia Research Forum here and read one of our first preprints here.
We are looking for an independent, self-driven Deep Learning Engineer to join our iSIZE team and contribute to transforming the world of streaming. Engage with world renown companies and help solve some of the biggest technical challenges behind advanced video systems that companies and environment is facing today. We are proud to be working with household names in the sector to reduce the strain on already-overburdened networks and cut the ballooning costs of the data centre and streaming industries, while improving the quality of end users.
- Develop and improve core deep learning architectures for video preprocessing for a variety of applications, such as live video streaming, Video-on-Demand, cloud gaming etc.
- Research and prototyping of new methods for fast image and video enhancement, particularly related to bandwidth saving and video streaming quality enhancement
- Draft publications and patent submissions in collaboration with other team members
- Participate and assist the team in commercial evaluations with top-tier companies
- MSc. in Computer Science, Electronic Engineering, Artificial Intelligence, Machine Learning or a related field, or equivalent skills evidenced by work experience in the specific domain of the post
- Strong background in: Python; TensorFlow (preferably v1) or PyTorch; evidenced by the development of advanced applications in image/video processing or computer vision (minimum of 12 months experience)
- Strong analytic background in deep neural networks, evidenced by knowledge of how to formulate and test advanced loss functions in neural network design, and design and test of advanced convolutional, recurrent or other neural network architectures in a task-specific manner
- Experience in training and inference with neural network architectures on CPUs/GPUs in cloud (AWS or GCE) and Tensorflow or PyTorch in VMs or Docker containers
- Experience in training, validation and evaluation of deep neural network models on large datasets, evidenced by experience in using Python libraries like h5py or similar
- Publications in top-tier conferences and journals: IEEE Transactions on Image Processing, IEEE Transactions on Circuits and Systems for Video Technology, conferences like IEEE CVPR/ICCV/ECCV, NeurIPS, ICML, ICLR, or similar
- Solid experience in image processing theory and methods, evidenced by the development of practical designs in this area
- Knowledge of image and video compression and the use of FFmpeg/libav libraries
For internship roles, the ‘minimum qualifications’ apply, with lower demands on experience.
Why join iSIZE team
- Be part of an award-winning & rapidly-growing start-up company in deep learning and video streaming technologies
- CDL Oxford AI Cohort graduate and backed by some of the leading tech experts
- Possibilities to interact with top-tier industrial labs who are trialling or using iSIZE technologies worldwide
- Collaborate and learn from professors and senior researchers who are part of the iSIZEteam
- Room to innovate and come up with breakthrough ideas
- Career evolution via publications and patent filings, which are strongly encouraged and are part of the official duties of the post
- Access and support to use state-of-the-art CPU and GPU cloud computing facilities
- Competitive salary and performance bonus
Apply by sending your CV, plus a cover letter, to firstname.lastname@example.org