Alejandro Koretzky

I am an innovator and problem solver at heart, with a strong Engineering & Applied Research background and an infinite supply of curiosity. I love sound and exploiting patterns in Data; most of all, I love building AI technology and products that inspire and advance the way people create.

As Head of AI at Splice, I lead the company's AI strategy and R&D division, building proprietary Machine Learning & Audio tech, co-driving Product vision, strategy and ideation for creation tools, while managing a fantastic team of researchers and ML engineers that I built from the ground up. I was the visionary behind products like Create (formerly CoSo), Search with Sound, Similar Sounds, Recommended For You and a co-inventor of the underlying technology, backed by a few granted patents. Currently, I am leading major investments in Generative AI to unlock new modalities for the future of audio editing and music creation. As a trusted advisor to the CEO and some of the Board members, I have had substantial influence and impact in shaping the company's direction since its relatively early days.

Before Splice, I was the founder of tuneSplit, an early-stage startup that developed pioneering real-time, deep learning based audio source separation technology (acquired in 2016). Before relocating to the U.S. from Argentina, I worked as a C/C++ Software Engineer at Motorola, led voice traffic prediction modeling for the largest credit card company in the country and co-founded tuquejasuma.com (StartupChile '13), today the leading complaint-resolution platform in Argentina.

My technical expertise combines machine learning, signal processing and software architecture, coupled with a strong intuition for human-computer interaction (I was a graphic/web designer a long time ago). My force-multiplying quality is orchestrating business goals & priorities, product vision & strategy, and hands-on technical leadership to create defensible value for a business and its customers, while fostering a culture of innovation and risk-taking within organizations.

My work has been widely recognized in the media, including Forbes, Yahoo, TechCrunch, Bloomberg, RollingStone, Billboard, DJ Magazine, Music Radar. I have been an invited speaker at top academic institutions, including Harvard Business School, Stanford University, Berklee College of Music, and organizations like Audio Engineering Society, SXSW, NAMM, among others.

Academic journey

I moved to the U.S. with a Fulbright Presidential Fellowship from Argentina and a Dean's Masters Fellowship from USC Viterbi, where I got my M.Sc. degree in Electrical Engineering, with a focus on Multimedia Signal Processing & Machine Learning. I worked with the Digital Signal Processing Lab, Immersive Audio Lab and Media Communications Lab, with advising professors -C Jay Kuo and Chris Kyriakakis. As part of my research work -which would later become tuneSplit-, I pioneered some of the early ML models for real time audio deconstruction, receiving incubation by the USC Office of Technology Innovation and Entrepreneurship and selected for an NSF grant.

Back in Argentina, I received my Engineer's Degree (5.5 years) in Electrical & Telecommunications Engineering, with a distinguished recognition for contributions to traffic prediction modeling and queueing systems (invited talks at Technion and Vrije University). During my fourth year, I received a full scholarship to study Computer Science at the University of Richmond, where I took courses on Data structures, Algorithms and Network protocols.

If I ever go back to school, I would consider Bioinformatics, Materials Science or why not my first love, Architecture :)

Advisory & consulting work

I believe in giving and helping others pursue bold ideas. Over the years, I've advised dozens of early-stage companies and organizations with Technology, Product strategy, Data and Hiring. I've been a mentor with Techstars (Roux Institute '22, Tel Aviv '24), StartupWeekend, the USC Startup Garage, advised Cornell Tech programs, consulted for YC-backed startups and occasionally help VCs and small funds with due diligence efforts.
Most memorable: consulting for SpaceIL as part of the Google Lunar X Prize competition. Most recent: Anyone AI.

Back in Argentina, I'm involved in a few Government-led initiatives to help develop the Public/Private tech investment ecosystem.

Music

I have a deep connection to music and music culture. I truly believe that music is the universal language and the social glue that keeps us in relative harmony with each other.

While doing Engineering at USC, I also attended the USC Thornton School of Music, where I studied songwriting with Andrea Stolpe and joined the Thornton Apollo Men's Chorus.
Back in Argentina, I was the producer, songwriter and lead singer of Ximetria, a rock band I started in 2006, with 2 studio album releases (Essential Innocence and Outliers) and music videos on MTV and VH1 Latin America. The band was a finalist at the Hard Rock Rising Buenos Aires 2012, played in London for the Youbloom Music Awards and got a couple of song placements on TV.

I am currently working on my next production project, 'Stolen Signals'. Expect some releases in 2025!

Some media coverage

A few pieces covering my work and profile. For a full list, click here

Selected talks, presentations & panels

  • Sparking Creativity: AI and the Future of Sound (2025) | SXSW '25
  • AI @ Splice: from Discovery to Creation (2024) | ISMIR 2024
  • Multimodal intent and curation-driven AI augmentation; new paradigms in creation (2024) | Harvard Business School
  • How a New Wave of Tools is Onboarding the Next 100M Music Creators (2024) | Mondo NY
  • International Symposium on AI: new paradigms in Music Creation. Opening talk (2024) | AES & Berklee College of Music
  • AI & Creative intent (2024) | FYI, will.I.am's creative space
  • Creative AI: what Artists and music industry professionals need to know (2023) | Water & Music
  • Inteligencia Artificial en la musica (2023) | Canal 12 Argentina
  • Rise of the Machines: Surviving the AI Music Apocalypse (2023) | NAMM '23
  • Exploring the Latent Spaces of Sounds (2023) | The Audio Programmer
  • Future of Music Summit LA (2023) | Skirball Cultural Center
  • The Future of the DAW: How AI will Change DAWs Forever (2022) | NAMM '22
  • Keynote: Signal Processing & Machine Learning: Audio applications (2021) | TechWeek Argentina
  • Applied Machine Learning for Audio Classification (2019) | Stanford University, CCRMA
  • Demixing and remixing music with Deep Learning (2017) | Stanford University, CCRMA
  • tunesplit: Semantic Equalization (2016) | Stanford University, CCRMA

Contact

You can reach me at 'alejandro' dot 'kor' at 'gmail' dot 'com'. You can also msg me over Linkedin