I’m Gargeya Sharma, an AI engineer and the founder of Edudojo.ai, with a background in theatre and computer vision. I build tools, write about learning, and make films when I travel.
As AI makes production easier, I keep coming back to what happens to the person doing the work. Are we becoming more capable, or just getting better-looking answers?
I’m building Edudojo to explore that question in education. Student work, questions, process journals, and feedback: a learning loop that values how someone gets there.
This website is the rest of the picture. The things I write to think more clearly. Films from places I’ve been. Small applications made out of curiosity. And the work as it develops.
I didn’t set out to work in AI. Follow the stops that brought me here. Choose any chapter.
School → university · 2018
Theatre was what I loved at school. My parents encouraged a more secure path, and I already liked science, so I enrolled in computer science, specialising in cybersecurity and forensics. It wasn’t my first choice. But I decided that if I was going to spend years on it, I owed it my best.
The pandemic · A campus in the mountains
Most students had gone home. I stayed near a campus surrounded by mountains, with very few people around and time to explore. I worked through more than a hundred online courses, doing the exercises as well as watching the lectures. Learning became something I chose every day.
Reinforcement learning caught me because I’d always been curious about psychology: here were machines learning from actions and feedback. I watched one course three times before Q-learning began to make sense. I wanted to keep going.
By graduation · 2022
Alongside my degree, I taught myself machine learning and deep learning, interned at a startup, and helped my professors with their research. That work led to a computer vision paper and a coauthored book chapter. I also wrote technical articles to help support myself. The subject I hadn’t chosen had become the thing I wanted to spend my days doing.
Read the publicationsQueen Mary University of London
Self-teaching had taken me a long way. For my MSc in Artificial Intelligence at Queen Mary, I wanted to find what I’d missed and learn alongside people who would challenge me. Moving from a cybersecurity degree into AI wasn’t a straightforward admissions path, but I got there. The coursework and dissertation gave me both a deeper foundation and more confidence in my research.
Explore the academic workLondon · Beyond the screen
Finding an AI role was harder than I’d hoped. Alongside the search, I worked in hospitality, from VIP lounges at The O2 to running a busy bar in Leicester Square. I learned to approach strangers, listen, manage a team, and stay composed when everything was happening at once.
Those conversations didn’t turn into the job I was looking for. They did change how I relate to people. After years of learning behind a screen, that mattered.
Now · Building
My own path involved choosing a subject, getting stuck, trying again, and finding people who could push me further. Those experiences shape the questions I bring to Edudojo: how can AI help someone develop understanding, and how can a teacher see that progress? I’m building it to make room for questions, revision, and the work behind an answer.
See what I’m building