Job Title:
Compression Researcher
Location: San Jose, CA, United States
Job Type: Full-Time
The world is experiencing explosive growth of digital content and
unstructured data. The accelerating growth of user generated content,
graphics intensive applications, and increased compliance requirements,
combined with the increasing sizes of most file types is straining
storage infrastructures worldwide. As a result, organizations are
struggling to contain rising storage acquisition costs as well as
operational expenses such as power, cooling and datacenter space.
A new solution is needed to optimize online content and reduce digital
footprint.
Ocarina Networks is funded by Kleiner Perkins
Caufield & Byers,
and Highland Capital Partners, and is tooled up to pioneer
the emerging file area network. Join our team and contribute to
this industry-changing effort!
Job Summary:
Ocarina is looking for someone to research, create, and
implement high-performance algorithms for Image and Video compression.
If you think you can develop algorithms to compress images and
video faster and more efficiently than anyone else, come do it
here.
Job Responsibilities:
- Research and implement cutting-edge compression algorithms
for video and images.
- Create better, faster, more efficient
algorithms using visualization and image analysis
and decomposition techniques.
- Implement the newly
created algorithms into future versions of Ocarina’s storage
optimization products
Qualifications:
- Successful track record of algorithm development for images
and video for commercial applications.
- Up-to-date knowledge
of the latest image detection techniques, lossy and
lossless compression methods, MPEG and JPEG-LS, PNG
algorithms.
- 5 or more years experience with arithmetic coding,
algorithm development, knowledge of techniques used
in PNG and JPEG2000, discrete cosine transform (DCT), and prediction
through partial match (PPM) based algorithms and image feature detection
techniques.
- A Ph.D. in Computer Science or higher mathematics is
required
Travel required: none
Telecommute: no |
|