CarbonTracker
Open research tooling that makes training-time energy and carbon costs visible for practical model development.
Pedram Bakhtiarifard
PhD Fellow @ University of Copenhagen, ML Section
bakhpedram@gmail.comCopenhagen, Denmark
About Me
A PhD Fellow in Sustainable ML at UCPH, working on algorithmic complexity theory, compression, and quantization as part of the SAINTS Lab. Current projects include algorithmic simplification of neural networks, both in theory and in practice. Previously led the development of CarbonTracker, recognized by the European Commission’s Innovation Radar, making the energy and carbon footprint of AI systems visible.
Earlier in industry, worked as a software engineer at Danske Bank, co-leading the end-to-end design and deployment of a large-scale mortgage refinancing system. It taught me a lot about performance and usability in real systems, which I bring with me into academia.
Outside the lab, I speak and write about AI’s rising compute demands and sustainable approaches. I have shares these perspectives at Thoughts for Future, D3A Conference, TV2 News, and DR Radio, as I believe research is most valuable when it is both directly useful and clearly understood beyond its immediate field.
Projects
Moments in time
A collection of photos from conferences, media appearances, research milestones, and everyday moments that I cherish.