Skip to content
View turancannb02's full-sized avatar
🇩🇪
🇩🇪
8 followers · 4 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Highlights

  • Pro

Block or report turancannb02

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.

Content in all repositories owned by your account will be closed.
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
turancannb02/README.md

Hi there, I'm Turan 👋

🔭 What I'm working on

  • Evaluating spec-driven agentic development (Spec Kit vs. OpenSpec): where coding agents help and where humans need to stay in the loop
  • Benchmarking AI coding agent configurations on reliability rate, tokens per success and turns to resolution
  • Fine-tuning small local models with LoRA (mlx-lm) for a personal assistant
  • Side builds: an iOS memory app and an ESP32 e-paper home dashboard

🌱 What I'm learning

  • Running coding agents in locked-down, sandboxed environments
  • Planning and reasoning strategies in agentic systems
  • LoRA fine-tuning and evaluation for small on-device models

💬 Ask me about

  • Measuring whether AI coding agents actually help, beyond vibes
  • Multi-agent LLM systems (built one for real-time network optimization during my MSc thesis)
  • Designing reproducible ML experiments
  • Turning research prototypes into working tools
  • Running local LLMs (Ollama, MLX) for practical tasks
  • Python data workflows and ETL pipelines

⚡ Fun fact

Built my first LLM agent on a 5G testbed. Still not sure which part was harder.

🌐 Socials

GitHub LinkedIn

💻 Tech Stack

Python PyTorch TensorFlow scikit-learn Ollama MLX FastAPI Pandas NumPy Plotly Matplotlib SciPy Databricks Apache Spark Delta Lake Azure DevOps GitHub Actions Jira Grafana SonarQube Pydantic Apache Airflow Docker SQLite Git GitLab Bash Script LaTeX nVIDIA

Pinned Loading

  1. opensre opensre Public

    Forked from Tracer-Cloud/opensre

    Build your own AI SRE agents. The open source toolkit for the AI era ✨

    Python

  2. techrag techrag Public

    Production RAG pipeline for technical documentation. Hybrid retrieval (FAISS + BM25), streaming FastAPI, Streamlit UI and deployed on HuggingFace Spaces.

    Python

  3. minigpt minigpt Public

    GPT-style transformer trained from scratch on TinyStories. PyTorch + BPE tokenizer + FastAPI serving. Built to understand LLMs from the ground up.

    Python

  4. bigdata-taxi-spark bigdata-taxi-spark Public

    Spark 3.5 notebook + Docker setup for NYC Yellow-Taxi trip-duration prediction (Big Data Analytics assignment)

    Python

  5. f1_qualifying_predictor f1_qualifying_predictor Public

    Machine learning–based Formula 1 qualifying lap time predictor using FastF1 telemetry and XGBoost.

    Jupyter Notebook 1

  6. unirag-ai unirag-ai Public

    Python