Yaman Umuroğlu
Principal FPGA Designer at EmLogic
Trondheim, Trøndelag, Norway
Summary
🤩
Rockstar🎓
Top SchoolYaman Umuroğlu is a Principal FPGA Designer and researcher with 13 years of experience bridging hardware–software co-design to accelerate deep learning and sparse graph algorithms. He pioneered the FINN framework for binarized neural networks and has contributed compiler and runtime features to Xilinx’s QNN dataflow toolchain, demonstrating rare expertise in extreme quantization on FPGAs. His roles across Xilinx, AMD and academia reflect a strong track record of turning research prototypes into deployable acceleration solutions and custom memory systems. Based in Trondheim and holding a PhD in Computer Architecture from NTNU, he combines hands-on RTL/FPGA design with ML systems thinking to squeeze efficiency from both hardware and models. A less obvious strength is his history of end-to-end embedded software work—from resource-constrained tablet clients to heterogeneous SoC acceleration—giving him practical insight into deployment constraints.
13 years of coding experience
10 years of employment as a software developer
BS Computer Engineering, BS Computer Engineering at Orta Doğu Teknik Üniversitesi / Middle East Technical UniversityHigh School Science, High School Science at Adnan Menderes Anatolian HighschoolBS Computer Science, BS Computer Science at Uppsala UniversityMaster of Science (MSc) European Master in Embedded Systems, Master of Science (MSc) European Master in Embedded Systems at University of SouthamptonNorwegian University of Science and Technology
English, Turkish, Norwegian, Swedish, German
Github Skills (15)
pytorch10
machine-learning10
hlsl10
onnx10
testing9
cprogramming-language9
c-language9
data-structure7
data-structures7
verilog7
algorithms7
neural-network6
hardware-designs6
deep-neural-networks6
python6
Programming languages (10)
Github contributions (5)
Xilinx/finn
Sep 2018 - Jan 2023
Dataflow compiler for QNN inference on FPGAs
Role in this project:
ML Engineer
ML EngineerContributions:6 releases, 20 reviews, 2445 commits in 4 years 4 months
Contributions summary:Yaman focused on developing and refining the dataflow compiler for QNN inference on FPGAs within the context of the FINN project. Their contributions involved working on various aspects of the compiler, including the implementation and integration of layers, shape inference improvements, and the integration of run-time writeable weights. The user's work is primarily centered around enabling efficient compilation and deployment of quantized neural networks for FPGA acceleration, as evidenced by their contributions to the test suite and integration of new Brevitas functionalities.
dataflow-compilerfpgasinferencecompilerneural-network
Xilinx/finn-base
Sep 2020 - Aug 2022
Open Source Compiler Framework using ONNX as Frontend and IR
Contributions:3 releases, 6 reviews, 125 commits in 1 year 11 months
finnonnxcompilerfrontendintermediate-representation
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.