Skip to content
View dimitrijepesic's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report dimitrijepesic

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
dimitrijepesic/README.md

Dimitrije Pešić

Software Engineering @ University of Belgrade

Interested in ML systems, compilers, GPU performance, and low-level software.

Research and project work includes LLVM pass ordering, MLIR compiler lowering, GPU workload optimization, and scientific machine learning.

Working with:

C++ · Python · LLVM · MLIR · PyTorch · TVM

Pinned Loading

  1. compiler-opt compiler-opt Public

    GNN-based LLVM pass ordering with PPO and GraphSAGE over CFG/DFG graphs for program-specific compiler optimization.

    Python

  2. mini-sql-query-optimizer mini-sql-query-optimizer Public

    Cost-based SQL query optimizer in pure Python. Takes a query and a JSON catalog with table statistics, picks the cheapest evaluation plan and prints the operation tree with the algorithm and cost i…

    Python

  3. JankoVukobratovic/VincaInternship JankoVukobratovic/VincaInternship Public

    Automated chemical vulnerability assessment of paintings from macro-XRF spectral imaging — U-Net denoising, element maps, NMF, CVI risk index, SAM segmentation (ICETRAN 2026)

    Python 2

  4. Pasado Pasado Public

    Forked from uiuc-arc/Pasado

    Research fork of Pasado: profiling and optimizing precise static analysis for neural-network robustness and GPU execution.

    Jupyter Notebook

  5. speech-recognition-2.0 speech-recognition-2.0 Public

    Benchmarking classical ML, CRNN, separable CNN, and attention models for spoken-digit recognition on FSDD.

    Python

  6. tvm-quantization tvm-quantization Public

    Compiler-level analysis of TVM quantization: int8 dispatch, low-bit failures, LLVM codegen, and cross-architecture performance.

    TeX