Open to opportunities

Hi, I'm

Arefeh Qavami

AI Engineer

I build the next generation of intelligent systems — from LLM-powered agents and multi-agent workflows to industrial computer vision.

LLM & Agentic AIComputer VisionNLPTehran, Iran
Years of AI Engineering
4+
Years of AI Engineering
Face Verification Accuracy
~98%
Face Verification Accuracy
MSc in AI, Ranked
2nd
MSc in AI, Ranked
About Me

Turning research into production AI

I'm an AI Engineer with a strong background in Machine Learning, Deep Learning, Computer Vision and Natural Language Processing, with hands-on experience applying modern architectures — including Transformers — to real-world vision applications.

My current focus is Large Language Models: Generative AI, Prompt Engineering, Agentic AI and Multi-Agent Systems, with practical experience in AI agents, tool calling, LangGraph, LangChain, MCP and Elasticsearch. I design AI workflows that integrate LLMs, tools, APIs, data retrieval and specialized AI services into reliable products.

Current focus areas

LLM & Generative AI

Prompt engineering, structured outputs and LLM-based summarization pipelines.

Agentic AI & Multi-Agent

Supervisor architectures with LangGraph, tool calling, MCP services and state management.

Computer Vision

Industrial detection and verification systems, from Siamese networks to YOLO.

NLP & Information Retrieval

Embeddings, semantic search, topic modeling and large-scale text clustering.

Technical Skills

A full-stack AI toolkit

From research models to production services — these are the tools I work with every day.

LLM & Agentic AI

  • Large Language Models
  • Generative AI
  • Prompt Engineering
  • AI Agents
  • Tool Calling
  • Multi-Agent Systems
  • Structured Outputs
  • LangGraph
  • LangChain
  • MCP

NLP & Information Retrieval

  • Text Embeddings
  • Semantic Similarity
  • Knowledge Retrieval
  • Topic Modeling
  • Text Clustering
  • BERTopic
  • Top2Vec
  • LDA
  • Jina Embeddings
  • Tokenization

Deep Learning

  • PyTorch
  • TensorFlow
  • Keras
  • CNNs
  • Transformers
  • ConvNeXt
  • Transfer Learning
  • Hyperparameter Tuning
  • Model Evaluation
  • Model Deployment

Computer Vision

  • Object Detection
  • YOLO
  • Image Classification
  • Face Verification
  • Siamese Networks
  • Image Processing
  • Template Matching

Machine Learning

  • Classification
  • Regression
  • Clustering
  • XGBoost
  • CatBoost
  • DBSCAN
  • HDBSCAN
  • K-Means
  • Hierarchical Clustering
  • Spectral Clustering
  • PCA
  • Scikit-learn

Engineering & Tools

  • Python
  • SQL
  • FastAPI
  • Flask
  • Django
  • Docker
  • Elasticsearch
  • MongoDB
  • MySQL
  • React
  • Streamlit
  • Async & Concurrency
  • Git
  • Linux
  • DVC
  • ONNX
  • Cursor AI
Work Experience

Where I've made an impact

AI engineering roles across agentic systems, robotics and smart-city technology.

AI Engineer · RCDAT

LLM systems, agentic AI and large-scale text intelligence.

Aug 2025 – Present
  • Built an end-to-end text analysis pipeline using Elasticsearch, Jina embeddings, clustering and LLM-based summarization, with targeted prompting and structured outputs for topic and emotion analysis.
  • Designed Supervisor-based Agentic AI architectures with LangGraph, enabling LLM-driven tool calling and request routing through specialized tools and subgraphs.
  • Developed MCP-based AI services that expose data-analysis and clustering capabilities as reusable tools for AI agents.
  • Implemented Elasticsearch retrieval, autocomplete and search suggestions, plus asynchronous LLM request handling with Python multithreading and Semaphore-based concurrency control.
  • Containerized AI backends with Python, FastAPI and Docker, and built frontend interfaces with React and Streamlit.
  • LangGraph
  • LangChain
  • MCP
  • Elasticsearch
  • FastAPI
  • Docker
  • React
  • Streamlit

AI Engineer · Nojan Robotics

Deep learning classification for an industrial pistachio sorting system.

Feb 2025 – Jul 2025
  • Prepared and augmented pistachio image datasets for industrial classification tasks.
  • Developed and trained transformer-based deep learning models, including ConvNeXt architectures.
  • Converted trained models to ONNX for efficient deployment on real-time R&D sorting devices.
  • ConvNeXt
  • Transformers
  • PyTorch
  • ONNX
  • Data Augmentation

AI Engineer · City Development and Innovation Company

Computer vision systems for smart-city and banking infrastructure.

Jun 2021 – Jun 2024
  • Developed a Siamese-network-based face verification system by fine-tuning pre-trained deep learning models, achieving approximately 98% accuracy.
  • Built national card detection with image processing and template matching (~93% accuracy) and a YOLO-based solution reaching ~95% accuracy.
  • Created CamQC — MongoDB queries and data-analysis workflows for quality control of ATM camera systems, in collaboration with a three-person engineering team.
  • Siamese Networks
  • YOLO
  • Template Matching
  • MongoDB
  • Quality Control
Education & Credentials

Academic foundation

Education

Artificial Intelligence — Master of Science

2017 – 2020

Alzahra University, Tehran

Ranked 2nd

Software Engineering — Bachelor of Science

2012 – 2017

Shahid Beheshti University, Tehran

Certificate

Advanced Computer Vision with TensorFlow

DeepLearning.AI · Coursera

Languages

EnglishFluent — TOEFL iBT 99 (2016)
PersianNative
Contact

Let's build something intelligent together

Whether it's an agentic AI product, a computer vision system or an NLP pipeline — I'd love to hear about it.

arefeh.qavami@gmail.com

Send me an email