Hello all,
CompSoc has had a great Michaelmas, and we hope you've enjoyed our events this term. If you have any feedback, please don't hesitate to contact a committee member :). Before I wish you a great vac, I'd like to slip in a few eighth week notices. Firstly, we have our Termly General Meeting on Wednesday, when the committee will report on the term and we plan to pass some minor changes to the constitution. If you can't come and you're a member, please contact me with your intention to vote by proxy.
We won't have an official geek night this Saturday, but if you're still in Oxford at least one committee member will be in the department on Saturday evening.
Have a great vac!
Thomas and the rest of the committee
Undergraduate Social Area, Department of Computer Science - 7pm Wednesday (8th week)
The committee will report on its activity this term, and we plan to pass a few minor changes to our constitution. In order to meet quorum, we need at least 20 people to attend. If you cannot attend but you are a member of the society, please inform a committee member of your intention to vote by proxy.
Food will be provided :).
Bloomberg HQ, London - 5:30pm Tuesday (8th week)
This event is in London, but Bloomberg have kindly offered to pay for train tickets for CompSoc members wishing to attend the event. If you're interested in going, please sign up via the above link and contact me so that I can pass on your name to Bloomberg if you'd like your ticket refunded.
Tech Talk: Data Mining usage patterns for building the right thing, building the thing right and supporting the thing better!
This talk will cover how Data Analysis and Machine Learning from usage patterns across large data sets is being used to predict bug reports, improve automated testing, and make tough decisions on what features to add to our software. We realised that data gathered for regulatory audit purposes, on millions of daily trades at Bloomberg, is an invaluable resource for analysis of our system's behaviour.
In this talk, we'll share the various opportunities this data has uncovered, the techniques we used for statistical analysis-based machine learning, and the data visualization behind it. The work we're planning and discoveries we've made promise to answer some very hard questions around prioritizing features to implement, reducing support costs, improving the automated test coverage and beyond. This also spawned an automated testing framework that allows the system to self-test by replicating the millions of daily trades in a secure test environment to achieve quality assurance. This has opened up horizons to build autonomic systems in the future systems that can both self-test and self-repair.
Trading Game
The talk will be followed by the Bloomberg Trading Game! This is a fun simulation where participants will be invited to understand the financial markets and trading activities through an active, sometimes frantic team trading game. Mentors will show you how to play the role of traders, sales and market makers and understand how these different players add value to the trading ecosystem.
This is a great opportunity to learn how Bloomberg's technologies help bring transparency and efficiency to the exciting and sometimes volatile financial market
Lecture Theatre A, Department of Computer Science - 1pm Tuesday (8th week)
25% of the worlds electricity consumption is consumed by a certain type of industrial machine, rotating equipment. An industrial plant can spend hundreds of millions of pounds annually operating these machines yet their operation is still highly inefficient. The operation of these machines is highly complex and their inefficiency is driven by human operators making ad-hoc decisions about their operation - usually based on the operators past experience. They have significant amounts of data at their disposal but it is far too complex for the human brain to comprehend on its own. In this talk we will demonstrate the problem at hand and show how Flexciton technology applies statistical models and mathematical optimisation to determine the optimal operation of these machines.
Jamie Potter, 25, is CEO and co-founder of Flexciton, a startup company disrupting the industrial world. He graduated from Oxford with a masters in Mathematics and Statistics and then began his career in a consultancy firm specialising in energy. There he built statistical software for several large companies including DECC, National Grid and RWE where his software was used to make multi-billion pound decisions.
EF headquarters, London - 4:30pm 8th December
On 8th December Entrepreneur First are hosting an engineering careers fair for startups in our portfolio who are looking to grow their teams.
Our alumni companies are some of the fastest-growing tech startups in Europe - they're building anything from low-cost robot arms, to deep learning visual inspection software; AI assistants for personal finance, to affordable nano-satellites. You can find out more about the companies attending here.
N.B. This careers fair isn't open to the general public and will be focused on technical positions only. Capacity is limited to 150 tickets, but if you'd like to reserve a free spot please RSVP.
If you would like to send across a CV ahead of the day to be submitted to all companies please email apply@joinef.com.
7pm Tuesday (8th week)
"What's an Estimathon" you ask?! It's a team contest where the goal is to create confidence intervals to difficult math and science questions. e.g., what's the volume of the earth's oceans (in cubic km); or, how many prime number contain strictly increasing digits.
It's a very interactive game and focuses on some ideas that are central to what we do at Jane Street: thinking about hard problems, assessing confidence levels, trying to strike a balance between quick-and-rough estimates versus more refined solutions.
There'll be prizes for the winning team and of course food and drinks will be provided.
Please sign up here
This December, we're excited to host our second edition of Women in Trading and Technology (WITT) at Jane Street. We're inviting women to spend two days in our London office to learn more about what we do and how we do it through a series of classes and activities. Attendees will receive an in-depth look into the ways we use math and computer science, as well as insight into the different roles that exist within our firm.
Selected students will arrive in London the morning of Thursday 8th December and will depart the evening of Friday 9th December. Jane Street will provide all travel to and from London, housing and of course food and drinks throughout the event.
The workshop itself will be an introduction to Jane Street. You will learn about how we use maths and statistics to decide what to trade and also find out how we use OCaml, a functional programming language, to build complex distributed systems. You will take part in some problem solving sessions and interactive games that will give you some insight in to how we work.
Application Process
Women interested in attending should apply here to be considered for a spot in the program. Soon after applying, you’ll receive an invitation to complete a puzzle.
The deadline to apply for this event is Wednesday 30th November. If you have any questions, please don't hesitate to email lborgo@janestreet.com
Good luck with the application process and we hope to meet you soon!
The Oxford University Computer Society (CompSoc) aims to organise meetings and events for our members to use and further their computing interests. See all of our upcoming events on our Facebook Page, Twitter, or visit our website for more information about the society.