Dragan Mladjenovic
Software Developer at Lanaco
Banja Luka, Republika Srpska, Bosnia and Herzegovina
Summary
👤
Senior🎓
Top SchoolDragan Mladjenovic is a pragmatic software developer with eight years of experience building backend and SaaS solutions from Banja Luka, Bosnia and Herzegovina. He brings strong expertise in ASP.NET MVC, C#, Web API and SQL Server reporting, and has a track record of delivering database-driven business applications at Lanaco. Beyond enterprise tooling, he contributes to high-profile open-source machine learning infrastructure—helping enable ROCm GPU support and low-precision convolution optimizations in TensorFlow—showing a rare bridge between traditional .NET systems and performance-sensitive ML engineering. His background in information technology from Elektrotehnicki fakultet Banja Luka underpins a methodical approach to software design and teamwork. Colleagues value his steady delivery, cross-domain curiosity, and ability to translate hardware-accelerated ML improvements into practical code.
8 years of coding experience
Bachelor's degree, Information Technology, 8, Bachelor's degree, Information Technology, 8 at Elektrotehnicki fakultet Banja Luka
English
Github Skills (11)
gpu-programming10
machine-learning10
roc10
tensorflow10
python9
cprogramming-language9
c-language9
deep-learning9
deep-neural-networks7
llvm7
neural-network7
Programming languages (6)
Github contributions (5)
tensorflow/tensorflow
Apr 2023 - Feb 2025
An Open Source Machine Learning Framework for Everyone
Role in this project:
ML Engineer
ML EngineerContributions:7 PRs, 3 comments in 1 year 10 months
Contributions summary:Dragan's contributions primarily focus on enabling and optimizing machine-learning related features within the TensorFlow framework, specifically related to ROCm (AMD GPUs). Their work includes enabling bf16 and int8 convolution for MIOpen, fixing build failures related to ROCm, and enabling fast f16 atomic add operations, indicating direct involvement in hardware-accelerated machine learning computations. They also added plumbing for hlo_to_llvm_ir tests and support for CUDA graph launch, further highlighting a focus on GPU-related performance and integration.
pythondata-sciencedeep-learningmlmachine-learning
draganmladjenovic/xla
Jun 2023 - Jul 2023
A machine learning compiler for GPUs, CPUs, and ML accelerators
Contributions:11 pushes, 1 branch in 21 days
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.