Applied AI & Machine Learning Engineer | Python, LLMs, AI Automation | UI Developer

I build practical AI systems end to end. My work runs from data exploration and model development to LLM-powered applications, backend services and APIs, automated workflows, and clean, responsive user interfaces.

I came to AI from geoscience. I hold a degree in Petroleum Engineering (Oil Exploration) from Imam Khomeini International University (IKIU). My research there used artificial intelligence and 3D seismic attributes to estimate reservoir porosity. That work taught me how to turn noisy, complex real-world data into reliable models people can base decisions on, and I bring the same approach to every project.

Today I work across machine learning, deep learning, natural language processing, and computer vision. I like taking a problem from raw data to a working solution: cleaning and analyzing data, building and evaluating models, explaining their behavior, and fitting them into real workflows.

To put models into production, I build backend services and REST APIs with FastAPI, Django, and Django REST Framework. That covers serving ML and LLM models through fast, well-documented endpoints, as well as building full Django backends with data models, authentication, and business logic.

I also design AI-driven automation workflows with Python and n8n. These workflows connect LLMs, APIs, and business tools so repetitive processes can run on their own.

My front-end development skills (HTML, CSS, JavaScript) let me build responsive, visually engaging interfaces with modern animations. Together with my backend skills, this means I can deliver the whole product: the model, the API that serves it, and the interface people use.

I also teach Python programming and artificial intelligence at the Shahriar branch of the Technical and Vocational Training Organization (TVTO) in Tehran. There I help learners move from fundamentals to hands-on AI projects.

Describe Image in Persian · Live Demo Upload an image and get a Persian-language AI description, with optional speech output. Built on GPT-4o vision and a Gradio interface, and deployed as a Hugging Face Space. Python GPT-4o Gradio Multimodal AI TTS

TTS Workflow An AI-powered text-to-speech automation workflow built on an OpenAI-compatible API, with a Gradio front end for easy use. Python Gradio OpenAI API Automation

Facial Emotion Recognition (FER-2013) A 5-layer CNN that recognizes human emotions from facial images. It uses data augmentation, max pooling, and dropout, and reaches 76% validation / 63% test accuracy on FER-2013. Useful for human-computer interaction. Python CNN Deep Learning Computer Vision

Handwritten Digit Recognition A CNN classifier for handwritten digits, written to PEP 8, clean-code, and SOLID principles. It includes a case study comparing the performance of simple and complex models. Python CNN Clean Code Model Comparison

Income Forecasting with Stacked Ensembles A stacked ensemble regressor that combines Gradient Boosting, AdaBoost, Bagging, Linear Regression, and Decision Trees. It reaches R² = 0.876. Python scikit-learn Ensemble Learning Regression

Sentiment Analysis on IMDB A Bernoulli Naive Bayes model that reaches 94% accuracy on IMDB movie-review sentiment classification. It shows how far a well-prepared simple model can go. Python NLP Text Classification NLTK

Energy Consumption Modeling Time-series forecasting of energy consumption. The project covers data preparation, feature engineering, outlier detection, and model evaluation. Python Time Series Feature Engineering Forecasting

• Applied Machine Learning: predictive modeling, feature engineering, model evaluation, and explainability
• Deep Learning: neural networks for vision and language tasks
• NLP & LLMs: text processing, Transformer models, and LLM-based applications
• Computer Vision: image analysis and recognition
• Backend & APIs: REST APIs and model-serving services with FastAPI, full backends with Django and Django REST Framework
• AI Automation: end-to-end workflows that combine LLMs, APIs, and webhooks
• UI Development: responsive, animated, user-focused web interfaces

Area	Tools
Languages
Machine Learning
Deep Learning
NLP & LLMs
Data & Visualization
Backend & APIs
Automation & Apps