Skip to content
View siwaarbn's full-sized avatar

Block or report siwaarbn

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
siwaarbn/README.md

Siwar Ben Nejma

Software engineer focused on AI/ML systems and developer tooling. Currently finishing my B.Sc. in Computer Science at TU Darmstadt, building LLM pipelines in production at a Frankfurt fintech, and teaching computer architecture as a TA.

I like working at the intersection of systems and intelligence — things that are fast, observable, and actually useful.

Projects

GROF — GPU profiler for AI workloads. Correlates CPU call stacks with GPU kernel execution via eBPF + CUDA, achieving <5% overhead vs. 10–50% for nsys/nvprof. Built the backend (FastAPI, PostgreSQL) and frontend (React, D3.js flamegraphs) as part of a 5-person team at TU Darmstadt.

codesearch — Semantic code search engine. Search any codebase by what functions do, not what they're called. Tree-sitter for AST parsing, sentence-transformers for embeddings, FAISS for retrieval.

ragbase — End-to-end RAG system for PDFs. Chunking, embedding, vector retrieval, grounded answers with source citations and RAGAS evaluation.

Stack

Python TypeScript React FastAPI PostgreSQL Docker eBPF CUDA FAISS ChromaDB

Contact

linkedin.com/in/siwar-ben-nejma · siwarbennejma2@gmail.com · Darmstadt, Germany

Pinned Loading

  1. codesearch codesearch Public

    Semantic code search engine — search your codebase by meaning, not keywords. Built with Tree-sitter, sentence-transformers & FAISS.

    Python

  2. grof grof Public

    Low-overhead GPU profiler for AI workloads — correlates CPU call stacks with GPU kernels using eBPF and CUDA

    Python

  3. ragbase ragbase Public

    Chat with any PDF using RAG — document ingestion, semantic search with ChromaDB, and Claude AI for grounded answers with source citations.

    Python