Open to opportunities

Karen Yepremyan

@karenyepremyan

Message

AI team lead and deep learning engineer specializing in production ML solutions and computer vision.

Russian Federation

Message

What I'm looking for

I seek a role leading ML/AI initiatives where I can apply production ML, computer vision, and MLOps expertise; I value clear customer collaboration, strong engineering practices, and opportunities to mentor teams.

I am an AI team lead and deep learning engineer with experience delivering production-grade enterprise AI solutions, computer vision systems, and generative models. I combined technical leadership with hands-on development—leading cross-functional teams, deploying RAG chatbots, building drone-based agronomy analytics, and implementing recommendation and ranking pipelines.

My background spans research and engineering across industry and freelance projects, including CLIP-based encoders, image matting publications, OCR pipelines, time-series anomaly detection, and scalable deployments on Kubernetes and AWS. I prioritize clear customer communication, robust MLOps, and measurable improvements in model relevance and system reliability.

Experience

Work history, roles, and key accomplishments

Current

AI Team Lead

Current

MWS (Mobile Web Services)

Feb 2025 - Present (10 months)

Led a cross-functional team of 3 data scientists and 2 engineers to deliver enterprise AI solutions on cloud and on-premises, including a RAG-based enterprise chatbot, computer vision agronomy analytics, and a CLIP-based product recommendation system that improved relevance and engagement.

RAG Chroma Kubernetes YOLOv8 Learning To Rank MLOps SQL

Deep Learning Engineer

Freelance

Nov 2020 - Dec 2024 (4 years 1 month)

Designed and deployed RAG pipelines, generative AI workflows, OCR systems, and chatbot solutions for legal, retail, and consumer clients, optimizing retrieval with DPR/FAISS and scaling deployments on AWS and Dockerized environments.

RAG Faiss Elasticsearch AWS EasyOCR PyTorch Docker Triton

Deep Learning Engineer

SberDevices

Sep 2022 - Aug 2024 (1 year 11 months)

Researched and developed textual and visual encoders (CLIP-based) and state-of-the-art image matting solutions, training SigLIP models on A100 GPUs and publishing a matte pulling network while improving portrait matting via synthetic augmentation.

PyTorch Image Matting

Full-stack Developer

ProvMe

Aug 2021 - Nov 2021 (3 months)

Built a social media application with a React frontend and Golang backend, implementing JWT auth, PostgreSQL (GORM), Redis caching, and Gorilla WebSocket for real-time messaging with performant concurrent handling.

Gorm React Go PostgreSQL GORM Redis JWT WebSockets

Data Engineer Intern

Joom

Aug 2021 - Nov 2021 (3 months)

Developed and deployed a scalable ML system for review ranking using Scala and Go, integrated Apache Kafka for real-time streams, and implemented Prometheus and Grafana for monitoring and alerting to ensure reliability.

Scala Go Apache Kafka Prometheus Grafana Performance Optimization

Education

Degrees, certifications, and relevant coursework

Moscow Institute of Physics and Technology (MIPT)

Master of Science, Applied Mathematics and Informatics

2021 - 2022

Completed a Master of Science in Applied Mathematics and Informatics with advanced studies and research in machine learning and computer vision.

Moscow Institute of Physics and Technology (MIPT)

Bachelor of Science, Applied Mathematics and Informatics

2017 - 2021

Completed a Bachelor of Science in Applied Mathematics and Informatics with coursework and projects focused on algorithms, machine learning, and software development.