Meta
Research Scientist Intern - Neural Representations Compression
Redmond, WA · Internship · Posted Oct 27, 2026
- Education
- Currently has, or is in the process of obtaining, a PhD degree in the listed or related field.
About the job
Research neural representation compression and streaming methods for split-inference systems, build and benchmark prototypes, and communicate research results.
What you'll do
- Explore processing, compression, and streaming of neural or learned representations; identify opportunities to improve compression efficiency or quality.
- Collaborate with peers to build a prototype algorithm.
- Communicate research progress and final results.
What they ask for
- Currently pursuing or obtaining a PhD in Image and Video Processing, Signal Processing, Information Theory, Machine Learning, Computer Graphics, or a related field.
- Coding experience in C/C++ and/or Python, plus experience in image and video processing.
- Track record of significant results, such as grants, fellowships, patents, or first-authored publications at leading workshops or conferences.
- Intent to return to a degree program after the internship or co-op.
Nice to have
- Knowledge of neural network and compression applications, image/video codec standards (H.26x, VPx, JPEG, etc.), MPEG PCC, or learning-based compression.
Pay
$7,650/month to $12,667/month
Tools
C/C++, Python, H.26x, VPx, JPEG, MPEG PCC
AI focus
Develop compression methods for neural representations, including latent vectors and Gaussian splats, to improve compression ratio, speed, and fidelity in split-inference AI and AR/VR systems.
Model-written summary · the original posting is authoritative