Avik Sengupta: Head of Engineering at Julia Computing Inc. at Cypher 2017
summary
Julia claims C-level speeds while staying high-level and dynamic, enabling models written by analysts to run in production without rewrites; Aviva reports 1000x speedup on Monte Carlo risk code and Celeste runs on 650,000 CPUs at petaflop scale across 30–40 TB of telescope images. Tools include the Flux deep learning stack, direct TensorFlow interop, and ability to compile to CUDA/Xeon Phi/CPU so the same Julia code moves from laptop to GPU or supercomputer and scales across clusters.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:41Why Julia
- 3:36Julia community
- 5:25Julia in industry
- 6:31FAA
- 8:00Use cases
- 9:51Machine learning
- 13:19Examples
- 13:51Blog Post
- 14:08Style Transfer
- 15:00HighLevel Language
- 18:57Julia Code
related talks
43mJulia: A Fresh Approach to Machine Learning
01h 27mIntroduction to Julia for Scientific Computing and Data Science
01h 28mA Tour of Julia
42mWhy I Use Julia For Quantum Physics
30mProductivity and performance in one programming language - A tour of Julia
01h 37mTech Tutorial: Use Julia for your next scientific computing job!
01h 06mJulia: Fast Performance, Distributed Computing & Multiple Dispatch
28mDiscover the Julia Machine Learning Ecosystem