Lilit Yolyan, Developer in Yerevan, Armenia
Lilit is available for hire
Hire Lilit

Lilit Yolyan

Verified Expert  in Engineering

Machine Learning Developer

Location
Yerevan, Armenia
Toptal Member Since
July 15, 2022

Lilit is an applied computer vision scientist with more than four years of experience. She's enthusiastic about solving real-life, everyday problems with computer vision. She works with an end-to-end AI pipeline from problem definition, data cleaning, and preparation to modeling, testing, and preparing the model for further deployment.

Deep LearningArtificial Intelligence (AI)Deep Neural NetworksPyTorchMachine LearningComputer VisionNatural Language Processing (NLP)CloudEnterprise Resource Planning (ERP)PythonOpenCVPyCharmSlackGitHubGitLabDall-E

Portfolio

SmartClick
Artificial Intelligence (AI), Computer Vision, Computer Vision Algorithms...
Yerevan State University
Computer Vision, Deep Learning, Artificial Neural Networks (ANN)...
BRAND BOUTIQUE LAWYERS PTY LTD Trading as Inkse
Google Cloud, Google Cloud Platform (GCP), Image Generation, Image Manipulation...

Experience

Image Segmentation - 4 yearsPyTorch - 4 yearsOpenCV - 4 yearsDeep Learning - 4 yearsPython - 4 yearsMachine Learning - 4 yearsComputer Vision - 4 yearsObject Detection - 4 years

Availability

Part-time

Preferred Environment

PyCharm, Slack, GitHub, GitLab

The most amazing...

...thing I developed was an anomaly detection model. I implemented paper by writing the code from scratch and then trained a model which works with high accuracy.

Work Experience

Machine Learning Scientist

2021 - PRESENT
SmartClick
  • Developed a product roadmap, established a hypothesis, and identified experiments to run.
  • Researched models, methods, and technologies that can solve real-life problems, including object detection, image segmentation, few-shot learning, image processing, etc. Implemented research methods and evaluated the results.
  • Improved the existing technologies by optimizing inference time and speed on real-life applications.
  • Worked with the team, brainstormed and discussed possible solutions, and helped team members overcome blocking issues and achieve end goals.
Technologies: Artificial Intelligence (AI), Computer Vision, Computer Vision Algorithms, Deep Learning, Deep Neural Networks, GitHub, GitLab, Image Classification, Image Processing, Image Segmentation, Machine Learning, Machine Vision, MongoDB, Neural Networks, Object Detection, OpenCV, NoSQL, Python, PyCharm, PyTorch, Self-supervised Learning, SQL, Object Tracking, Team Leadership, TensorFlow, Slack, Cloud, Kubernetes, Artificial Neural Networks (ANN), Generative Adversarial Networks (GANs), Convolutional Neural Networks (CNN), 3D Image Processing, 2D, Data Science, OCR, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), GPT

Lecturer

2019 - PRESENT
Yerevan State University
  • Created a syllabus for two different classes, Deep Learning and Large Scale Analytics.
  • Managed a group of teacher assistants to work with me on courses.
  • Held lectures and practical classes for 15–25 students.
Technologies: Computer Vision, Deep Learning, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), 2D, Data Science

Back-end Developer with Image Manipulation | Online Startup

2022 - 2023
BRAND BOUTIQUE LAWYERS PTY LTD Trading as Inkse
  • Created two PoCs to generate prints for clothing.
  • Documented the process and technology of the two PoCs fully.
  • Prepared reports and presentations for weekly meetings.
Technologies: Google Cloud, Google Cloud Platform (GCP), Image Generation, Image Manipulation, Generative Adversarial Networks (GANs), Artificial Intelligence (AI)

AI Developer

2022 - 2022
Mohammed Ehsan Ur Rahman
  • Fixed bugs and errors in GAN regarding training loops, data loaders, and network architecture.
  • Experimented with different network architectures to get the best results on the given dataset.
  • Optimized the code for faster performance and adhere to PEP8 standards.
Technologies: Computer Vision, Python 3, Generative Adversarial Networks (GANs), TensorFlow, Deep Learning, PyTorch

Machine Learning Scientist

2020 - 2021
SmartClick
  • Worked on end-to-end processes, including data collection and preparation, cleaning, modeling, testing, and error analysis.
  • Implemented computer vision techniques such as classification, object detection, segmentation, and metric learning for solutions in various domains. Modified and adapted models for specific solutions.
  • Collaborated with the engineering team for model deployment.
Technologies: Image Classification, Object Detection, Image Segmentation, Computer Vision, Deep Learning, Machine Learning, Python, PyTorch, OpenCV, Object Tracking, Computer Vision Algorithms, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), 2D, Data Science

Data Scientist

2018 - 2020
Menu Group
  • Developed machine learning models for customer behavior prediction.
  • Automated reporting processes for sales and management teams.
  • Created end-to-end machine learning pipelines for different products.
Technologies: Python, Machine Learning, Convolutional Neural Networks (CNN), 2D, Data Science, Marketing, Enterprise Resource Planning (ERP)

CutPaste Model Implementation

https://github.com/LilitYolyan/CutPaste
CutPaste is a self-supervised learning algorithm created by Google. I implemented the code from scratch on GitHub by writing the model, training, and data loading codes. I also reproduced the results of the paper.

Development of Few-shot Learning Technology

I've developed an end-to-end few-shot learning technology that works with images of money, stamps, and other flat objects. The purpose of the model is to detect an object in the given images, compare it with the database, and output the most similar image. I trained the model and developed the database structure and API code to bring it together.

3D Deep Learning with Python

I am the co-author of the book "3D Deep Learning with Python," which will be released soon. It includes a theoretical and practical description of various applications of 3D computer vision using Python.

Libraries/APIs

PyTorch, OpenCV, TensorFlow

Paradigms

Data Science

Other

Deep Learning, Image Classification, Object Detection, Image Segmentation, Artificial Intelligence (AI), Image Processing, Machine Vision, Neural Networks, Deep Neural Networks, Computer Vision Algorithms, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), 2D, Machine Learning, Computer Vision, Team Leadership, Self-supervised Learning, Cloud, Object Tracking, Point Clouds, Generative Adversarial Networks (GANs), 3D Image Processing, OCR, Natural Language Processing (NLP), Enterprise Resource Planning (ERP), GPT, Generative Pre-trained Transformers (GPT), Few-shot Learning, Voice Cloning, Text to Speech (TTS), Image Generation, Image Manipulation, Stable Diffusion, DALL-E

Languages

Python, SQL, Python 3

Tools

PyCharm, Slack, GitHub, GitLab

Storage

MongoDB, NoSQL, Google Cloud

Industry Expertise

Marketing

Platforms

Kubernetes, Google Cloud Platform (GCP)

2017 - 2019

Master's Degree in Data Science

Yerevan State University - Yerevan, Armenia

FEBRUARY 2022 - PRESENT

Nvidia Deep Learning Institute Certification

Nvidia Deep Learning Institute

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring