Lean Confectionary & Java Bootcamp for Quants

A packed house once again for the June meetup with plenty of faces old and new. Recently there seems to be more people arriving well ahead of the start time of the first scheduled speaker and taking the opportunity to grab a bite to eat and network or, as it should be more correctly called given we are in Yorkshire - scran and chelpin. The evening followed the typical format of a 30 minute talk followed by 60 minute one.

#####Andy Stewart: Learn Software Development in 8 weeks Andy Stewart kicked off the evenings presentations with an experience report for a project where financial quantitative analysts ('Quants') were directly involved in the implementation of financial algorithms in Java.

This was despite the Quants having no previous software development experience beyond MS Excel spreadsheets. The team had 2 senior developers, 3 quants and 2 functional testers and the project was judged by senior management to be an unqualified success with early delivery, low defect rates and 8 week ROI. The success of the project was put down to a number of factors: senior developers and quants pair programming extensively and directly in Java, zero tolerance to defects and coding standards violations, a comprehensive test suite with the focus on a large number of unit tests, support in the code for easily creating test data and very fast build times of less than 1 minute.

The fast build time was achieved by only having a small number of automated system tests. In addition, infrastructure detail was abstracted away so the dev/quant pair could focus more on the problem domain. The talk was well received and timed to allow plenty of opportunity for questions of which there were many.

#####Dom Hodgeson: An Agile Sweet Shop? After the break the presentations continued with Dom Hodgeson's account of taking an idea from it's (alcohol inspired) beginnings to the current day. Like all good accounts it was full of ups and downs with excitement and frustration in equal measure.

The starting point for the journey was a hypothesis: can you do pick and mix online? On the way he discovered that free postage to Norway is expensive, people get passionate about sweets, civil engineering infrastructure projects do not move at lean startup pace and users sometimes can't differentiate between a squirrel and a monkey.

What we learned is that we need to ask the question "what is the worst that can happen?" and that the first thing to do, even before a business plan, is to buy a domain name. The talk generated a huge amount of questions and discussion which spilled over into the post-meeting session in the pub - which was made even more pleasant by the warm summer evening allowing a large number of us to set up in the outdoor terrace of the Midnight Bell.

The answer to the original hypothesis "can you do pick and mix online?" seemed to be a resounding yes if you learn from your mistakes and keep asking what is the worst which can happen. Jelly snakes anyone?