Mücahit Yılmaz

An AI engineer working across computer vision and LLMs — four years of shipping vision systems to production, alongside RAG pipelines and agentic systems.

Get in TouchView projects

About Me

I’ve spent nearly four years building computer vision systems that run in production — real-time object detection, tracking, OCR and anomaly detection. I take projects from dataset design all the way to edge deployment on Jetson with TensorRT and ONNX, integrated with industrial cameras and PLCs.

I write clean, maintainable Python (OOP, SOLID) and get the architecture and data pipeline right before touching the model. I’ve delivered a commercially licensed product, led a project that has run in production for over a year without retraining, and I mentor junior engineers and interns.

Alongside computer vision, I work on large language models — RAG pipelines, agentic systems and fine-tuning — and bring the same production mindset to them.

Technical Skills & Expertise

The stack behind my LLM work and production computer vision systems.

LLM & RAG · Backend & MLOps

  • LangChain
  • LlamaIndex
  • FAISS
  • OpenAI API
  • Autogen
  • Transformers
  • PEFT · LoRA/QLoRA
  • Gradio
  • Python
  • FastAPI
  • Docker
  • MongoDB
  • PostgreSQL
  • Azure DevOps
  • Linux

Computer Vision · Edge & Deployment

  • PyTorch
  • TensorFlow
  • OpenCV
  • YOLO
  • RT-DETR
  • SAHI
  • ByteTrack
  • PaddleOCR
  • MediaPipe
  • Albumentations
  • TensorRT
  • ONNX
  • CUDA
  • NVIDIA Jetson
  • PyQt6
  • Industrial cameras
  • PLC integration

Selected Projects

All projects →

Road Damage Detection — Production Fleet

From a publicly funded R&D prototype to a production fleet system deployed to a metropolitan municipality, with a 10-vehicle rollout in progress.

  • 87%detection accuracy
  • 10vehicles in rollout

GIP — End-to-End Object Detection Platform

A standalone desktop platform covering the full object detection lifecycle — camera capture, annotation, training and real-time GPU-accelerated inference — in a single application.

Solar Panel Quality Control

Eliminated missed-cell errors in cell counting and improved small-object detection with SAHI. Delivered as a subscription-based, server-licensed commercial product.

  • 95%+small-object detection accuracy

Rob Ros — Cobot Drawing System

An end-to-end cobot application that turns photos or line art into physical pen drawings with a Lexium cobot, with a safety-interlocked desktop HMI for industrial demos.

LLMs, RAG & Agents

Independent projects across retrieval, agents and fine-tuning.

RAG-Based Knowledge Assistant

A RAG pipeline over sample documents using a FAISS vector database and LangChain / LlamaIndex to answer queries grounded in the source.

LangChain · LlamaIndex · FAISS · OpenAI API

Agentic AI Systems

A multi-modal support assistant routing text, voice and image inputs through tools, and a multi-agent workflow for retrieval, analysis and reporting.

LangChain · Autogen · Function calling · Gradio

Summarization & Fine-Tuning

An audio-to-summary tool comparing frontier and open-source models, and a price-prediction model fine-tuned with LoRA/QLoRA.

Transformers · PEFT · LoRA/QLoRA

GitHub Activity

486 contributions in the last year.

Work Experience

Ithinka Information Technologies

Jul 2022 – Present · 3 roles

  1. Apr 2023 – Present

    Computer Vision & AI Engineer

    Own end-to-end delivery of computer vision projects: dataset design, model training and optimization, edge deployment and production monitoring. Work directly with clients from requirements through go-live, including on-site installation and end-user training. Mentor junior engineers and interns — assigning and tracking tasks, reviewing work and onboarding them onto production projects — and produce architecture, setup and handover documentation for every delivered project. Take a research-first approach, reviewing the academic literature before adopting new techniques in production, and worked in Scrum sprints during periods of larger team size. Early on, delivered dataset and model-training work on industrial quality-control projects such as End-of-Line Inspection and Spring Sorting.

  2. Oct 2022 – Mar 2023

    Data Scientist, Data Team

    Built an anomaly detection system for abnormal network traffic on monitored servers, including balancing a highly imbalanced dataset, and presented weekly progress reports and analysis recommendations to stakeholders.

  3. Jul 2022 – Sep 2022

    Data Science Trainee

    Completed a 3-month data science traineeship at Ithinka before being hired, covering foundational data analysis and machine learning workflows.

Education

  1. Erciyes University

    2015 – 2018

    B.Sc. Geomatics Engineering

Certifications

Contact

Have an LLM, RAG or computer vision project in mind — or something that needs to go from prototype to production? I’d be glad to hear about it. I work remotely.

Message sent

Thanks for reaching out — I’ll get back to you soon.

Prefer email?

Or let’s connect on