I help with getting our AI prepared for clients and questions. An AI has to be fed with lots of data, so it can learn and make better decisions. That is what I and my teammates are doing. We will then need to test the AI after we trained it, to ensure it answers questions as we would think it does. Next, we need to check what it is doing when actually confronted with real people. 🙂
In my current job, I don’t do much AI at the moment, I did before but now less. As part of my PhD I personally favour more robotics than AI. AI is mostly a software that finds things efficiently among a lot of options, robotics is all about moving things (most often a chain of iron bars arranged like an arm) based of what sensors have found and what the robot has been commanded to do. There is an overlap between AI and robotics, as a robot might still need to find the best path to reach a door or an object. As such, I did help progressing AI applied to robotics but only what it is useful, some things I had to leave them back (i.e. responding in English to a question asked in a microphone, which one of the classical AI jobs)
Most of my work is about AI and how it can help people work better together. I work a lot with mathematics and logic to make that happen, and I build programs that use mathematical rules to help make decisions.
Comments