-
Asked by on 18 Jun 2020. This question was also asked by .
-
Lucy Craddock answered on 18 Jun 2020:
I’ve been working for Airbus for 2.5 years and in that time I’ve had three projects. The first two involved using natural language processing (an area of artificial intelligence that enables a computer to classify and analyse text). I’m not sure what more information I can share on that, sorry! The last and current project is using artificial intelligence to approximate lengthy and complicated physic calculations for flight design. This allows designers and engineers to run quicker programs to assess their aircraft designs.
-
-
anon answered on 18 Jun 2020:
I’ve worked on many over the years. Perhaps one of the most memorable was when I worked for NatWest Bank back in the 1990s. We pioneered “electronic cash” on chip cards, under the name “Mondex”, and my team was involved in securing it, including running a special room in one of our data centres, built as a Faraday Cage (completely encased in electrically-conducting metal cladding, so no radio waves could get in or out), where new electronic money was “minted” with special dispensation from the Bank of England.
-
Katy Brown answered on 18 Jun 2020:
My first project was studying monkeys and other primates, looking for evidence of ancient viruses which infected them millions of years ago.
Then I took a break from working on viruses and worked on three different projects – one about the causes of kidney disease, one about evolution of genes which make some people more likely to get cancer and one about how hearts develop in embryos.
Now I’m back working on viruses again, I study the different viruses which infect bees, ants and wasps, how they evolve and how they are passed between different species.
-
Martin Coath answered on 18 Jun 2020: last edited 18 Jun 2020 12:06 pm
I have been working for quite a few years 🙂 But in the last three years I have been working on climate data and writing apps which allow people to see and make use of the predictions. This is quite straightforward from a programing perspective – the big challenges lie in other apects of the project.
Before that I worked in a field called ‘neuromorphic engineering’, where engineers design and build silicon chips that work in an unconventional way, inspired by the brain. My job was to examine what such a chip might be able to learn, and to write simulations of the learning that would tell us what would happen in a real chip.
Before that I worked in Neuroscience where I was interested in how we hear, and how we learn to hear.
I did some interesting work during my time studying: I looked at using AI to learn a musical style and to compose original tunes in the same style, I did some modelling of how particles behave in a mass spectrometer. Loads of other small things I won’t go on for ever! 😉 -
Fabian Grosse answered on 18 Jun 2020:
The projects I’ve been working on are all around direct and indirect impact of human activities on the marine environment using ocean models. In the past, I’ve studied how excessive inputs of nutrients from rivers affect growth of algae in coastal seas and what the (mostly negative) consequences of this are. Currently, I am working on the impact of the decline in sea ice in the Arctic (due to global warming) affects algal growth and colleagues of mine look into how this affects animals further up in the food web.
-
Will Furnell answered on 18 Jun 2020:
I’ve worked on a few cool projects over the last few years, some at school and some at work!
At school I worked with radiaton detectors that let you see different radioactive particles! They’re used at CERN and in some schools that are lucky enough to have them! I helped write some software to analyse data that we got from these.
At work, I’ve previously done a six month project on updating some of our monitoring system. We have hundreds of servers, and need to know whether they have any problems with them at any time. I did some coding for this, to automatically monitor the servers without manually inputting every single one.
My current project is related to a system that lets you log onto lots of different websites via a single site with a single username and password. I’m programming a dashboard to help manage all the different people that log in, and to make it as easy as possible for someone to sign up to the website. I’m also looking at ways researchers can access our servers via this site too.
-
Steve Williams answered on 18 Jun 2020:
The project I worked on that received the most publicity was the Remove Debris mission which was launched from the International Space Station. I designed and wrote the software that sent instructions to the cameras and then downloaded the images and videos off them. This was an experimental mission and was the first of it’s kind to investigate two experimental methods of capturing in orbit satellites which where no longer in use – often referred to as “space junk”. The experiment used two methods, a net and a harpoon to determine how they would behave in zero gravity. The net experiment was used on a target satellite where it managed to successfully seize the craft. The next stage of the experiment is tow the captured spacecraft back into the earth’s atmosphere where it is safely burnt up upon re-entry.
The BBC covered this story and showed the video taken from the remove debris spacecraft of the experiments in operation.
-
Freya Addison answered on 18 Jun 2020:
I am early in my career but have been lucky enough to work on a few different projects. PICASSO, is a project (University of Manchester led) that was using the UK’s atmospheric research aircraft (FAAM https://www.faam.ac.uk/), to make new observations using cloud imaging probes of ice in clouds, and compare them to ground based radar & additional weather balloon measurements. I was working on checking that the observations made were of the same parts of the clouds, but as a student I got to learn about different aspects of the work, including calibration of some instruments and flying on the aircraft. I did a bit of work on a project called CONCIRTO (Université de La Réunion led) which is looking at cirrus clouds over Reunion Island, mainly calibration based. I have gotten more and more involved in a project called BioDAR (University of Leeds led) (https://biodarproject.org/) where we are seeing if we can see insects using weather radar and whether we can obtain any further morphological data (what types of insects can we see). My PhD aspect is part making sure we are observing the same space, and part the amount of biological mass there is. My current work on BioDAR is modelling insects and running simulations to see what they might look like from the radar perspective.
I also worked on a merging galaxy classifier before it was launched on Galaxy Zoo (Zooniverse project https://www.zooniverse.org/) as a Nuffield Foundation Research Placement (for 17yo). -
Emma Wilson answered on 18 Jun 2020:
All the projects I work on focus on how we can use a method called systematic review to identify, evaluate, and summarize the findings of all relevant studies over a field of research, thereby making the available evidence more accessible.
So far I’ve worked on reviews of stroke and Alzheimer’s disease. More recently, I’ve started a systematic review project related to COVID-19, where I get to work with scientists from the World Health Organisation.
In all the projects I do I need a key set of skills:
1) knowledge of the subject matter (I have a degree in Neuroscience which helps me review stroke and Alzheimer’s studies)
2) library/librarian skills (for searching for and finding all the relevant studies)
3) data science skills (to analyse results, and develop automation processes to help me perform some tasks) -
Lauren Baddeley answered on 18 Jun 2020: last edited 18 Jun 2020 4:42 pm
As a Software Engineer, I get to work on lots of projects that last from as little as 6 weeks to some that last 6 months to longer than a year.
In my first job at GE Aviation Systems, I got to work on the Electrical Load Management System, which was the next generation of the power management systems for the Boeing 777X aircraft. It manages the power that goes to all the electrical parts on a plane so things like the light above your head, the TV screen in the headrest, the fuel management system – all of these things are controlled by code! I was involved with requirements gathering, model design and managing our global teams from the UK, Poland and Mexico. Also at GE I got to work on in-house tools that other software engineers needed for their daily job. One such tool was a Verification Tool for Boeing 777X aircraft software.
Now I work for a company called Amiosec, a UK technology company specialising in the design and development of secure solutions for government and commercial organisations. I now work in the cyber security industry. I’ve been involved in working on our main product which allows remote access to sensitive networks. On this project I’ve been the Scrum Master (which facilitates and organises the team and their workload), I’ve developed code (in C, C#, Python etc.) and I’ve got to meet with customers to find out what issues they are having.
What is great about being a software engineer is that there are so many industries you can work in – I’ve worked in aviation with planes, I now work in cyber security – but maybe in the future I will work in automotive, in space, in defence, in retail – who knows! The opportunities are endless as everyone needs a software engineer these days.
-
Sreejita Ghosh answered on 19 Jun 2020:
During my masters internship I used a Machine learning algorithm to try and identify the functional MRI scans of the brains of patients in the early stages of schizophrenia and those in the later stages of schizophrenia. Back then I was infatuated with neuro-imaging. This project introduced me to Machine learning and AI and since then I have been in love with interpretable machine learning.
For my Masters thesis I used an interpretable machine learning method called Learning vector quantization (LVQ) to identify different stages of heart failure, and also identify the more important factors which corresponded with the more riskier group of heart failure patients. -
Anar Yusifov answered on 20 Jun 2020:
I was Windows programmer using Delphi, wrote few search engines, data mining and anomaly detection systems. Then was cryptography engineer developing my own signature algorithm and embedding it into Excel (oh boy). Software Project and Product manager. Then tired of all of that and by accident found myself to be a Linux developer, writing algorithms of polygon blending, interpolation using wavelets and FFT, 2D spectral modeling and it all in good old Fortran. Then gradually came back to my Soft Computing background with Evolutionary Algorithms and Neural Nets. And from that countless number of projects of different sizes and different backgrounds – from Data Science to JavaScript and from gradient based optimizations to deep learning.
-
Nikita Moghe answered on 22 Jun 2020:
I mostly work on projects related to making computers understand language. One of them was about building a movie chatbot that talks to you about your favorite movie and also suggests a similar movie based on the chosen movie.
I also like a project from my undergrad where we built Food Smiles – an initiative to prevent food wastage at the domestic and city level. For the domestic level, we built a recipe engine that provided a list of recipes depending on the ingredients which are in your fridge, essentially, the left-overs. For the city level, we built a portal that connected restaurants that had excess food and NGOs that can collect this excess food and distribute them to people who are homeless and hungry.
Of course, that was a basic model but the thought behind the effort was appreciated 🙂 -
Laura Murgatroyd answered on 22 Jun 2020:
At university, my final year project involved looking at, and simulating data from the ALICE (https://home.cern/science/experiments/alice) experiment at CERN. ALICE stands for A Large Ion Collider Experiment. It is used to study collisions of lead ions at the Large Hadron Collider (a 27km long particle accelerator). Scientists are interested in studying these collisions because when these heavy ions collide at such high energy, they can create a state called a quark gluon plasma. This is similar to a state that existed in the first fractions of a second when the universe was formed!
Physicists want to understand more about the formation of the universe, and to do this they want to study what this quark gluon plasma is like. In the particle collisions in the ALICE experiment, sometimes there will be “jets” of particles. Jets are just clusters of high energy particles. If we can identify these particles and measure how they are affected by travelling through the quark gluon plasma, then this can tell us about the properties of the plasma.
This may be tricky to understand, but think about having a glass of water and a glass of an unknown liquid. You could drop a metal ball into your glass of water and time how long it takes to sink to the bottom. Then you could do the same for the unknown liquid. If the ball took longer to fall through the mystery liquid then you may conclude that the unknown liquid is more viscous (stickier and harder to travel through) than water. This is kind of similar to how studying a jet can tell us about the properties of the quark gluon plasma.
However, the collisions of the lead ions are very messy – they result in LOADS of particles being produced. We call the big mess of particles we aren’t interested in the “background”. This makes it hard to find the jets. My project was about comparing different methods of finding the jets amongst this background.
I studied lots of data from real collisions, and used that to make a random simulation of a typical “background”. What I could then do is create a jet with known properties that I set myself (such as its speed) and add this jet into the background.
Then I tried out using various different methods to subtract off the background. My goal was to find the best technique to do this. I did this lots of times with many jets and compared the results. A perfect technique would have left me with a jet with the exact properties that I started out with. As well as testing currently used methods, I tried out one that I coded myself. I actually found it to be as effective as the current methods in some cases, even though it was much simpler!
Comments
David commented on :
I have worked on lots of exciting projects with many amazing people. As a production chemist it was part of my job to solve problems that impacted on oil production. One exciting project was to investigate the use of a new chemical (Acrolein) on the removal of a sludge we called “Schmoo” in the waterflood network. This involved installing special pipes which allowed the chemical to be injected and using computer data from sensors to monitor the effectiveness of the program. It worked very well and we were all happy with the results!