About
Since my teenage years until the end of my twenties, I lived a dual life. Both art and engineering were so fascinating to me that I couldn’t give up either. Alongside writing plays and directing theater, I earned my Bachelor's and Master's degrees in Computer Science from Amirkabir University. On top of that, I worked as a programmer. Even though it was tough, I’m very happy because pursuing my Master's in Music Technology at Georgia Tech placed me on an academic path where both worlds finally merged.
If you ask my friends what Farshad does the most, they'll tell you I’m a writer. Writing began for me in my teenage years, when I would search through my grandfather’s old books for inspiring stories and always chose to write stories instead of reports during composition classes.
For my thesis in Music Technology, I created a multimodal film music dataset and web-based explorer, built around a large collection of feature films with aligned musical, textual, and visual features. You can probably guess how excited I am about it, as it combines literature, cinema, music, and computer science. Creating datasets is something I find truly enjoyable, perhaps rooted in my childhood admiration for my grandfather, who collected magazine archives. For my Computer Science master’s thesis, I published a dataset of Persian vocal music in MusicXML format, and for my undergraduate project, I built a tool for creating a multilingual machine translation dataset. Following that path, I’m now converting the Radif, the most important book for teaching Persian music, into symbolic music formats.
In August 2025, I graduated from my Master’s in Music Technology at Georgia Tech and started my PhD in Music at the University of Cambridge, in the Centre for Music and Science, supervised by Dr Peter Harrison. My doctoral studies are supported by an AHRC Open-Oxford-Cambridge DTP fellowship. In my research, I am using the film music dataset I developed in my master’s as the foundation for extracting richer musical, textual, and visual features, studying cross-modal patterns, and exploring how musical meaning emerges in multimodal ensembles.
I now see myself staying in academia. After years of working as a software engineer, what I saw in that future was a rich but unsatisfied employee. I prefer to look at the positive side of that experience, which has given me the flexibility to take on almost any project. My work with diverse programming languages has made learning new tools and systems feel natural. Additionally, my long-term, in-depth experience with distributed systems serving tens of millions of users is invaluable for scaling academic projects for public use and broader impact. My time as a tech lead in large software companies has also familiarized me with contemporary, efficient organizational cultures, making me well-suited to lead young, academic teams.
These days, I’m thinking a lot about new artistic mediums. Cinema was the first step in changing my perspective after a decade of working in theater as a writer and director. In recent years, I’ve produced three experimental short films, often taking on key roles such as screenwriter, director, cinematographer, and sound designer, sometimes out of necessity and sometimes out of passion. Parallel to this, I’ve started a personal project: an audio drama studio built on a multi-agent LLM architecture, where different agents handle story structure, dialogue, directing decisions, and the sonification of events. In the longer term, I aim to add user interaction, turning these systems into truly immersive and interactive experiences. I feel that, just as cinema in the 1920s transformed from ambitious ideas into a new art form that no critic could have fully predicted, today we have tools that can evolve into the next medium. I hope that one day I can explore these new forms with these two key concepts in mind.