Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/ML Engineer with 3+ years of experience building and deploying production-grade machine learning systems, computer vision pipelines, and full-stack applications. I enjoy translating ambiguous, real-world problems into reliable, scalable AI-powered features—from data preparation and model training to serving, monitoring, and continuous improvement. I’m hands-on with PyTorch and modern deep learning architectures (CNNs, U-Net, Transformers, SegFormer), plus end-to-end MLOps workflows and SQL-based data engineering. I also build the product layer when needed, including backend APIs and React frontends, to deliver complete, measurable solutions.…AI/ML Engineer with 3+ years of experience building and deploying production-grade machine learning systems, computer vision pipelines, and full-stack applications. I enjoy translating ambiguous, real-world problems into reliable, scalable AI-powered features—from data preparation and model training to serving, monitoring, and continuous improvement. I’m hands-on with PyTorch and modern deep learning architectures (CNNs, U-Net, Transformers, SegFormer), plus end-to-end MLOps workflows and SQL-based data engineering. I also build the product layer when needed, including backend APIs and React frontends, to deliver complete, measurable solutions.

Samane Vazirian

AI Developer, Database Developer, Data Analyst, +5





AI/ML Engineer with 3+ years of experience building and deploying production-grade machine learning systems, computer vision pipelines, and full-stack applications. I enjoy translating ambiguous, real-world problems into reliable, scalable AI-powered features—from data preparation and model training to serving, monitoring, and continuous improvement. I’m hands-on with PyTorch and modern deep learning architectures (CNNs, U-Net, Transformers, SegFormer), plus end-to-end MLOps workflows and SQL-based data engineering. I also build the product layer when needed, including backend APIs and React frontends, to deliver complete, measurable solutions.…AI/ML Engineer with 3+ years of experience building and deploying production-grade machine learning systems, computer vision pipelines, and full-stack applications. I enjoy translating ambiguous, real-world problems into reliable, scalable AI-powered features—from data preparation and model training to serving, monitoring, and continuous improvement. I’m hands-on with PyTorch and modern deep learning architectures (CNNs, U-Net, Transformers, SegFormer), plus end-to-end MLOps workflows and SQL-based data engineering. I also build the product layer when needed, including backend APIs and React frontends, to deliver complete, measurable solutions.

Available to hire

AI/ML Engineer with 3+ years of experience building and deploying production-grade machine learning systems, computer vision pipelines, and full-stack applications. I enjoy translating ambiguous, real-world problems into reliable, scalable AI-powered features—from data preparation and model training to serving, monitoring, and continuous improvement.

I’m hands-on with PyTorch and modern deep learning architectures (CNNs, U-Net, Transformers, SegFormer), plus end-to-end MLOps workflows and SQL-based data engineering. I also build the product layer when needed, including backend APIs and React frontends, to deliver complete, measurable solutions.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Beginner

Work Experience

AI/ML Research Engineer at CERC Smart Cities Lab, Concordia University

January 1, 2024 - Present

Designed and implemented Python-based MLOps workflows covering data preparation, feature extraction, training, fine-tuning, model comparison, inference, and evaluation to enable reproducible AI development. Prepared and transformed large-scale urban datasets to ensure pipeline reliability and data quality. Built computer vision pipelines for semantic segmentation to predict building floor counts from street-level images, handling the full ML lifecycle from raw data to inference outputs. Developed frontend updates in React and improved platform structure to support data-driven urban analytics. Trained and evaluated CNN, U-Net, LSTM, Transformer, and SegFormer models for multiple tasks including semantic segmentation and predictive analytics. Working on a scalability challenge for bulk data import and slow JOIN performance across millions of records via indexing, query optimization, and aggregate-table strategies. Implemented a RAG chatbot using LLMs, embeddings, and vector database retr

AI/ML Research Engineer at CERC Smart City Lab, Concordia University

January 1, 2024 - Present

Estimated building floor counts from street-level imagery with 77% accuracy and within ±1 floor accuracy. Built and evaluated deep learning CV models (SegFormer, U-Net). Developed an end-to-end ML pipeline (data preprocessing, fine-tuning, evaluation, inference, JSON result generation). Containerized and deployed models with Docker and AWS. Proposed integration plans to update missing floor-count attributes in urban databases via facade-image based models. Built a context-aware question answering system over domain-specific datasets using RAG-based pipelines, LLMs, embeddings, and vector databases. Implemented a hybrid Transformer–LSTM model for energy consumption forecasting on time-series data. Trained/evaluated transformer-based and CNN models for image classification and optimized configurations. Improved map tile loading performance for Colouring Cities by optimizing data workflows, cleaning datasets, updating React frontend components, and enhancing backend APIs.

Software Developer at Shokouhejavid

January 1, 2021 - January 1, 2023

Accelerated reporting performance for high-volume financial data by ~30% through optimized SQL queries, aggregation workflows, and summary-table pipelines. Reduced database/server load by implementing scheduled ETL and data-processing jobs for large-scale transactional datasets. Enhanced system monitoring/observability with Elasticsearch-based logging solutions. Improved data integration across platform services by developing and maintaining backend REST APIs.

Full-Stack Web Developer at Sepas Rayaneh

January 1, 2018 - January 1, 2021

Automated/manual processes for questionnaires, university registration, and transportation workflows by developing and maintaining full-stack web applications. Designed scalable database architectures and application logic to improve performance across multiple data-driven products. Collaborated with technical and business stakeholders to deliver reliable solutions across projects.