Friday Hacks #195, October 16

Posted on by Chaitanya Baranwal

Date/Time: Friday, October 16 at 7:00pm
Venue: Online on Zoom
Zoom Link:

Online Algorithms Made Easy

Talk Description:

The world around us is constantly changing and at Jane Street we’re receiving new data at a rate faster than ever before. When running complicated models that contain many parameters, reacting to a change in a single parameter by restarting the entire model is impractical. Instead, Jane Street makes use of ideas from research into self-adjusting computation in order to turn static models into models that can be partially recomputed in an efficient way. This talk will introduce the idea of self-adjusting computation and show how it can be used to make online algorithms easy to run and update efficiently.

Speaker Profile:

Jane Street is an electronic market maker with offices in New York, London and Hong Kong. Harry graduated from Victoria University of Wellington, New Zealand with a BSc (Hons) in Computer Science and is now a Software Engineer in Jane Street’s Hong Kong office working on a distributed system at the core of Jane Street’s trading infrastructure.

Our Journey with Kubernetes, One Year in Production

Talk Description:

This session will give a very brief introduction of containers and Kubernetes, and share our experiences in introducing Kubernetes to our infrastructure, covering both the good and bad. At the end, Jiang Huan will also share some tips for university students and fresh graduates to start learning Kubernetes with free online resources.

Speaker Profile:

Jiang Huan has 4 years of career experience in software development, specialized in backend development, cloud infrastructure automation and Kubernetes. He currently works as a Senior Research Engineer in Titansoft’s Research & Development Team, where the team is focusing on furthering the company’s technical direction and driving innovations across the stack. He is also a fan of functional programming and video games.

See you there!

comments powered by Disqus