What is AI Hack

The largest student-organised data science hackathon in the UK. Our flagship AI Hack this year will be a virtual 24h frenzy of hacking and data science. It aims to bring the most forward-thinking and creative student data scientist to solve some of the world's most pressing challenges. Joining us for the hackathon will be a superb experience to

Test your data science skills and to apply them to real projects
Chance to level up your CV for the career goals you aspire towards!
A chance to win part of our £2000 worth of prizes
The experience of obtaining professional mentorship during our event

Form teams, Solve REAL projects and win prizes.

Key Dates

Applications Open
Applications Close

Schedule

Opening Address
Keynote Speech- Dr Rossella Arcucci
Project Presentation- Dr Sibo Cheng
Hacking starts
Freebies on campus!!
Workshop on ML! Joshua Mills
Talk: Energy Price Forecase & Optimisation - Yuya Ra (Limejump)

Talk: What makes a good pitch? - Bryan Liu
Hacking Stops & Submission Deadline on DevPost
Presentations from top 3 teams on the two challenges
Closing Address- Event ends

Challenges

Two challenges. Hint: one will focus on image data while the other will be on tabular data.

Judges

Dr Rebecca Pope (she/her) Rebecca is the UK's Digital & Data Science Innovation Lead and Strategic Theme Lead in Personalised Healthcare at Roche. Rebecca's focus is working with the NHS, governments, regulators, clinicians and patients on how we can bring the advanced technologies that we’re used to using in our everyday lives, like artificial intelligence, into healthcare. She is also a honorary Senior Research Fellow and PhD supervisor in AI at UCL, TEDx speaker and has been voted as one of 'The Most Influential Women in UK Tech'. Rebecca and is deeply passionate about equality, diversity & inclusion in STEM & society and, is part of several initiatives & networks that encourage an inclusive environment where everyone feels empowered.

Dr Nausheen Mehboob Basha (she/her) I graduated with a PhD in computational fluid dynamics modelling of compressors. Currently, I work as a project manager for a program aimed at advancing machine learning and computational fluid dynamics to create ultra-fast predictive models. I also research on optimisation of chemical reactors through data-driven techniques. Outside of work, I enjoy jigsaw puzzles, gardening and reading articles on Medium.

Harrison Zhu (He/him) Harrison is a 3rd year PhD student at Imperial College London, supervised by Seth Flaxman and Yingzhen Li. His research is focused on developing statistical machine learning methodology with applications to climate science and public health. He has also participated in many hackathons before, such as the Citadel Data Open, where his team was ranked 3rd place globally.

Bryan Liu Bryan wears two hats – he works part-time as a Senior Machine Learning Scientist at ASOS.com and part-time towards a PhD at Imperial College London & University of Oxford. In both roles he is interested in improving the controlled experiment design and causal inference methods for digital experimentation & measurement, enabling decision-makers to make less biased decisions more quickly.

Suwen Ge Suwen is a software engineer at Twitter's Ads Infra team. Before joining Twitter, he worked at AWS Database Migration Services for two years, building features to optimize migration speed for NoSQL databases. At Imperial, he was an active member of Imperial College Data Science Society and held the position as the president in 2017-2018.

Dr Sibo Cheng Dr. Sibo Cheng is a research associate at Data science institute of Imperial College London. He works mainly on prediction of dynamical systems using the combination of machine learning , reduced-order-modelling and data assimilation techniques. His research interests also include network science and application of data-driven approaches in engineering problems.

Robyn Lucas Head of Data Science and Forecasting at Limejump.

Robyn Lucas Satya works as a Hydrogen Portfolio Analyst at Shell. He focuses on green hydrogen, where his engineering background from Imperial College London meets his interest in commercialisation. He is building advanced analytics tools to measure, track, and optimise hydrogen value chain. He is interested in developing data-driven business by applying technology to improve the quality of life.

Prizes

Project 1 Winner

Congratulations!
Project 1 Runner ups x2

Cash voucher of £25 per person, use it to buy anything you desire.
Project 2 Winner

Congratulations!
Project 2 Runner ups x2

Cash voucher of £25 per person, use it to buy anything you desire.

Frequently Asked Questions

AI Hack is Imperial College Data Science Society's annual datathon where students get to showcase their analytics skills and meet like-minded aspiring data scientists.

AI Hack will be run mostly virtually this year and will take place from 12th of March — 13th of March. Join from anywhere to hack and listen to our talks!

There will be some merch and also some free food for all participants who signed up for the in-person tickets.

To be able to produce a good submission, we would recommend some familiarity with some of the following things:

Data and software engineering: pandas, PyTorch, TensorFlow, Tableau, parallelisation, docker, GCP or AWS, ipython…
Data visualisation: dplyr, ggplot, matplotlib, seaborn, plotly
Machine learning: Nonparametric smoothing, random forests, finite and infinite mixture models, XGboost, kernel methods, Gaussian processes, neural networks, autoencoders…
Statistical modelling: frequentist and Bayesian linear models, frequentist and Bayesian hypothesis testing, Monte Carlo methods, bootstrap simulations, PCA…

If you're starting out in your data science journey and want to use this to gain some experience, that's completely fine!. You can focus on data visualisation, applying models such as linear regression, logistic regression and decision trees and also doing hypothesis testing.

We recommend Python, but you could also use R, Julia or any language that can be used for scientific computing.

Teams consist of 3-4 people

The Discord channel set up before the Hackathon itself for all individual sign ups to meet new people and teammates.

Sadly we won't be able to accommodate people who miss the sign-up date.