Ahmet Özlü

5.0

(227 reviews)

US$20.00

For every 15 mins

295

Sessions/Jobs

View Video

ABOUT ME

Computer Vision Expert

About Me

I am an experienced AI and Computer Vision Specialist with a Master’s degree in Computer Vision and a recipient of the Google Inside Look 2019 Award. My expertise lies in developing advanced solutions in Deep Learning, Machine Learning, and Image Processing, with a strong focus on building and optimizing AI-driven systems.

I have successfully built AI teams using Gemini, Groq, and Phidata, enabling multi-agent collaboration for tasks such as data collection, processing, and analysis. Additionally, I specialize in Face Recognition, Object Detection, and Video Analysis, leveraging technologies like TensorFlow, Keras, and OpenCV to deliver impactful solutions.

Key Achievements

Optimized an object detection pipeline for a startup, reducing processing time by 40%.
Built and deployed end-to-end ML models for image classification in production environments.
Guided junior developers in designing and training custom CNNs, improving team performance.

Mentorship Philosophy

I excel at breaking down complex AI concepts into digestible steps, empowering both beginners and experienced developers. Whether you're starting your journey in AI or looking to fine-tune and scale your solutions, I’m here to guide you with practical, hands-on advice.

GitHub: github.com/ahmetozlu
Papers: Facial Expression Recognition | Automatic Age Estimation

English

Istanbul (+03:00)

Joined October 2018

EXPERTISE

Computer Vision

5 years experience | 127 endorsements

I have expertise on object detection/tracking/counting, facial image analysis, augmented reality and image segmentation. I have been deve...

I have expertise on object detection/tracking/counting, facial image analysis, augmented reality and image segmentation. I have been developing the Computer Vision projects using both modern approaches via CNN models with TensorFlow and dlib and traditional Computer Vision algorithms via OpenCV and scikit-image. I have been developing REST API based cloud solutions for Computer Vision tasks such as cloud base face verification systems. My papers are about face recognition and computer vision: - https://ieeexplore.ieee.org/document/8404767/ - https://ieeexplore.ieee.org/document/8404549/

RESTful API Linux TensorFlow OpenCV Python C++Ubuntu Flask NumPy Pandas

Deep Learning

4 years experience | 126 endorsements

I have been designing&training DNN models with TenworFlow - Keras API. I have expertise in transfer learning, fine-tuning and training ap...

I have been designing&training DNN models with TenworFlow - Keras API. I have expertise in transfer learning, fine-tuning and training application specific CNN models from scratch! I have taken cs231n lecture: http://cs231n.stanford.edu/syllabus.html and I got all the details it has. I use OpenFace frameworks for face recognition&verification based systems. I have been developing "TensorFlow Object Counting API" which is an open source framework built on top of TensorFlow that makes it easy to develop object counting systems and it has many stars on GitHub: https://github.com/ahmetozlu/tensorflow_object_counting_api I have been developing object detection/tracking/counting systems, face recognition&verification (OpenFace) systems and image segmentation tasks. Moreover, I have been working on time series analysis via Deep Learning techniques. I can develop any Deep Learning based requirement and integrate it with the cloud structure. Also I have very strong theoretical background of the Deep Learning!

Linux Ubuntu TensorFlow

Linux Ubuntu TensorFlow Keras OpenCV Python C++NumPy Pandas Matplotlib

Machine learning

4 years experience | 128 endorsements

I have a quite strong theoretical and practical background in Machine Learning. I have been implementing and developing the projects with...

I have a quite strong theoretical and practical background in Machine Learning. I have been implementing and developing the projects with the most popular Machine Learning classifiers such has Support Vector Machines, and with the most popular data preprocessing techniques such as LDA&PCA and so on. I have excellent background in application specific training for classification&regression based Machine Learning models with the features are extracted from the desired data! I have been integrating my implemented and developed Machine Learning classification&regression models to the cloud systems for analyzing and interpreting the real time data!

Python Linux Ubuntu

Python Linux Ubuntu NumPy Pandas Matplotlib C++

LLM

2 years experience

I specialize in designing, building, and mentoring teams to develop and deploy LLM (Large Language Model) Agent Systems. My expertise inc...

I specialize in designing, building, and mentoring teams to develop and deploy LLM (Large Language Model) Agent Systems. My expertise includes structuring and coordinating multi-agent AI systems using advanced tools like Gemini, Groq, and Phidata. I guide teams in implementing real-world applications of LLM agents, which involve tasks such as data retrieval, task automation, and intelligent system orchestration. Additionally, I develop workflows that integrate LLM-powered automation, data pipelines, and custom APIs to address complex business problems. I also coach junior developers on leveraging LLMs for tasks like content generation, data analysis, and interactive system design. I excel at fostering collaboration and innovation within AI teams, ensuring that each agent is tailored to meet specific project needs while maintaining scalability and reliability. Whether you're building your first LLM agent team or optimizing an existing one, I bring proven strategies and hands-on experience to drive success.

Large Language Models MLOps Neural Networks

Large Language Models MLOps Neural Networks NLP LLM AI Google gemini Langchain OpenAI Huggingface

Object-Oriented Programming

5 years experience | 113 endorsements

I have been developing web based projects mainly with REST API. I have been using spring boot (java) and web API (c#) for developing RE...

I have been developing web based projects mainly with REST API. I have been using spring boot (java) and web API (c#) for developing RESTful web services. I have been using Swagger as a documentation tool. Using Maven as dependency management tool, git for version control, Jenkins for Continuous Integration and JBoss as deployment tool.

RESTful API Hibernate ORM Spring Boot

RESTful API Hibernate ORM Spring Boot Java SwaggerAsp. net, c#Jenkins Git JBoss

Python

5 years experience | 119 endorsements

I have been developing most of my projects on python. I have expertise in algorithm development, data mining, computer vision, machine le...

I have been developing most of my projects on python. I have expertise in algorithm development, data mining, computer vision, machine learning, REST API (flask) and so on.

TensorFlow OpenCV Machine learning

TensorFlow OpenCV Machine learning Computer Vision Deep Learning Flask NumPy Pandas Django

Java

6 years experience | 114 endorsements

I have expertise in developing back-end services and applications on java. I love spring boot framework and also love coding in java!

RESTful API Spring Spring Boot

RESTful API Spring Spring Boot JDBC JBoss Spring Security Spring data JUnit Java

REVIEWS FROM CLIENTS

5.0

(227 reviews)

Redox berr

October 2022

Greatly explained! During the session i was provided with very on point and detailed information. I surely recommend his services to anyone.

Redox berr

October 2022

Great work as usual!

Redox berr

October 2022

Amazing work and delivered lighting quick. I recommend his services to anyone whose interested in quality and reliable work

Monty Washington

April 2021

Excellent help from Ahmet, He was able to help with my issue very quickly.

Jayden Sherif

January 2021

Ahmed is great!

Jayden Sherif

January 2021

This is one of many sessions I had with Ahmet, he is simply great, especially with the way he explains difficult concepts . He is very experienced in the fields of Deep Learning, Machine Learning, computer vision and anything that has to do with python. I definitely recommend him, he is great!

Jayden Sherif

December 2020

I had my first session with Ahmet, he is great and extremely knowledgeable when it comes to Deep learning and computer vision!

mohammed

November 2020

not totally satisfied with the lesson! but I learnt one thing even thought I didn't get what I am looking for

evan picow

October 2020

Ahmet Is very knowledgeable in Pandas, Keras & Tensorflow. Very easy to work with.

evan picow

October 2020

Ahmet is very knowledgeable in machine learning. He's also easy to work with and enthusiastic.

SOCIAL PRESENCE

GitHub

tensorflow_object_counting_api

🚀 The TensorFlow Object Counting API is an open source framework built on top of TensorFlow and Keras that makes it easy to develop object counting systems!

Python

1297

540

vehicle_counting_tensorflow

:oncoming_automobile: "MORE THAN VEHICLE COUNTING!" This project provides prediction for speed, color and size of the vehicles with TensorFlow Object Counting API.

Python

852

363

Stack Overflow

1255 Reputation

EMPLOYMENTS

Senior software engineer

Garanti BBVA Technology

2018-02-01-Present

- Design, lead development and maintenance of Online & Bulk SMS System of Garanti Bank (BBVA) which serves millions of users every day, a...

- Design, lead development and maintenance of Online & Bulk SMS System of Garanti Bank (BBVA) which serves millions of users every day, and work on AI based research projects that have critical revenue impact. - Using Maven as dependency management tool, git for version control, Jenkins for Continuous Integration and JBoss as deployment tool.

Java

JBoss

Jenkins

Java

JBoss

Jenkins

Spring Boot

Swagger

RESTful API

Computer Vision Researcher

SiMiT Lab

2017-07-01-Present

- Researching novel facial image processing and analysis, under the supervision of Prof. Hazim Kemal Ekenel. - Developing multi-view fac...

- Researching novel facial image processing and analysis, under the supervision of Prof. Hazim Kemal Ekenel. - Developing multi-view face recognition system by my own design CNN architecture to improve accuracy and speed of current face recognition systems.

Python

Linux

OpenCV

Python

Linux

OpenCV

Ubuntu

NumPy

Matplotlib

Pandas

TensorFlow

Keras

Software Research and Development Engineer

Erarge

2016-07-01-2018-01-01

- I involved in C3PO (Collaborative City Co-Design Platform), an EUREKA ITEA3 Project where 21 partners from Turkey, Finland and Belgium ...

- I involved in C3PO (Collaborative City Co-Design Platform), an EUREKA ITEA3 Project where 21 partners from Turkey, Finland and Belgium participate. - I was responsible from Pendik/Istanbul case study, where the effects of urban transformation modeled and presented by using AR & VR, IoT, Cloud Technologies, led to receive "EUROKA Innovation of Tomorrow Award" and "ITEA Award of Excellence".

PHP

Python

Android

PHP

Python

Android

Java

C++

SQL

MongoDB

PROJECTS

TensorFlow Object Detection APIView Project

2018

The TensorFlow Object Counting API is an open source framework built on top of TensorFlow that makes it easy to develop object counting s...

The TensorFlow Object Counting API is an open source framework built on top of TensorFlow that makes it easy to develop object counting systems!

Python

NumPy

Machine learning

Python

NumPy

Machine learning

Computer Vision

Deep Learning

TensorFlow

Collaborative City Co-design PlatfOrm (CP3O)View Project

2017

C3PO aims at providing a Cloud collaborative and semantic platform for city co-design. The C3PO platform is unique in that it covers the ...

C3PO aims at providing a Cloud collaborative and semantic platform for city co-design. The C3PO platform is unique in that it covers the whole urban project development process where cities empower, encourage and guide different stakeholders to develop an urban project together. C3PO does not intend to replace or modify the existing applications offering unique but partial solutions of city co-design but can be seen as an open and generic intermediary that enables the interaction between existing applications through a unique multi-dimensional semantic repository. Awarded "EUROKA Innovation of Tomorrow Award" and "ITEA Award of Excellence" in 2018.