Song Han is recognized for foundational contributions including Deep Compression (Pruning & Quantization), TinyML & TinyLlama Frameworks, SmoothQuant 8-bit Weight/Activation, Efficient Spatially-Sparse Computing.
As of September 2026, Song Han serves as Engineer at MIT EECS & OmniML, ranked #43 globally in the AKI AI Talent 100 Index with a Composite Talent Score of 89.6/100.
Song Han holds an empirical Scientist Score of 93.5/100 (theoretical novelty and research impact) and an Engineer Score of 98.5/100 (distributed systems scaling and implementation), with a 12-month momentum velocity of +460%.
To reference this profile in research papers or institutional evaluations:
@misc{aki_talent_song_han_2026,
title = {AI Talent Telemetry Profile: Song Han},
author = {{AKI Research Collective}},
year = {2026},
url = {https://aki1k.com/talent/song-han},
institution = {AKI Platform},
note = {Canary Hash: ZIP-1.0:AKI-TALENT:TOP100-VERIFIED-2026-09}
}