Netflix Video Algorithms Intern Fall 2026 | Gaussian Splatting

Overview

Gaussian Splatting (GS) enables photorealistic novel-view synthesis with low rendering complexity — attractive for TVs, streaming sticks, phones, and laptops. The open challenges: much faster model training/encoding and more efficient compression.

On the Video Algorithms team during this 24-week Fall internship, you’ll investigate GS as a future streaming format and build toward a practical system: compression strategies on open datasets, size/training-time/quality trade-offs vs streaming-rate targets, experiments to cut encoding time or improve compression, and a proof-of-concept GS rendering pipeline on content of interest.

Key Requirements

• Eligibility: PhD student in CS, Engineering, Math, or Statistics with expected graduation June 2027 or later.
• Research fit: Track record in 3D/4D scene reconstruction, novel-view synthesis, Gaussian Splatting or NeRF, differentiable rendering, neural graphics, or 3D computer vision.
• Skills: Strong software engineering practices (VCS, testing, code review); solid ML/DL fundamentals with hands-on model training/eval; fluent Python.
• Nice to have: Real-time rendering / GPU (CUDA, WebGL); video compression or codecs (HEVC, AV1); open-source multimedia/graphics; large-scale distributed systems / cloud.
• Why it stands out: Long 24-week research internship on next-gen streaming; embed with Netflix Video Algorithms; publish-adjacent work on GS for consumer devices.