Arian Naseh

A demo web app that leverages large language models to screen resumes against job descriptions and provide recruiter-style recommendations.

• Problem: Traditional resume screening is manual, slow, and inconsistent.
• Solution: Built a scalable app that takes in a resume and job description, uses OpenAI GPT to analyze alignment, and returns a recommendation with rationale.
• Tech Stack: Streamlit, LangChain, OpenAI GPT-3.5, Pydantic, Document Comparison, GitHub-based deployment

A web-based ML app that predicts whether a user would be approved for a loan based on their financial profile.

• Problem: Users and financial institutions need quick, explainable credit risk assessment tools.
• Solution: Developed a user-facing app that predicts loan eligibility using a trained XGBoost model and displays results instantly.
• Tech Stack: Streamlit, Python, XGBoost, Docker, Heroku, CI/CD pipeline, Medium Blognd.
• Fully containerized with deployment to cloud registry
• Includes user input validation and basic explainability
• Accompanied by detailed blog walkthroughs

• Leading ML efforts focused on predictive modeling, LLM-based automation, and cloud-native infrastructure.
• Developed XGBoost classifiers and regressors to predict auto insurance losses, optimizing risk rankings and saving over $1M annually.
• Built ETL pipelines using Snowflake, AzureML, and Blob Storage for seamless data integration and transformation.
• Designed and deployed a scalable FastAPI service for automated First Notice of Loss (FNOL) claims processing using OpenAI GPT models and LangChain.
• Established CI/CD workflows with Docker and Azure Container Registry for production-grade deployments.
• Implemented model retraining pipelines with automated hyperparameter tuning using Hyperopt.
• Created forecasting models using AutoARIMA, CatBoost, TSB, Croston, and Prophet, securing a $500K annual resource expansion approval.
• Championed GitHub Copilot adoption to boost developer productivity through internal surveys and feedback.

• Designed and optimized API endpoints for targeted advertising analytics, improving execution efficiency and data reliability.
• Enhanced audience segmentation models while preserving statistical fidelity, leading to improved marketing insights.
• Collaborated with the team to refine code repositories for better maintainability and streamlined onboarding processes.
• Worked on backend development to improve campaign performance and profiling algorithms.

• Developed ML models to optimize plant growth prediction for indoor farming, improving yield forecasting accuracy.
• Created real-time dashboards using Python, SQL, and Power BI to monitor environmental metrics and alert operators to suboptimal conditions.
• Designed a data-capturing Android application to streamline data collection workflows for farm operators.
• Delivered actionable insights by visualizing sensor data and recommending adjustments for better productivity.

• Co-founded a financial technology startup focused on democratizing algorithmic investment strategies.
• Built and deployed portfolio selection systems using Python, integrating technical and fundamental financial indicators.
• Designed ETL pipelines to handle large-scale historical financial data, ensuring accuracy and consistency.
• Developed modules for validating portfolio back-tests, enhancing strategy robustness and decision-making.
• Led marketing campaigns and contributed to shaping the initial business model.

Completed a Master of Science in Computer Engineering with a focus on Machine Learning and Data Science. Relevant coursework included Machine Learning Theory, Probabilistic Models, Data Analytics & Visualization, and Data Mining. My thesis explored the application of Deep Reinforcement Learning in optimizing caching strategies for IoT networks, achieving significant reductions in energy consumption.

Pivoted from Electrical Engineering to Computer Science, with a strong focus on Statistical Machine Learning and Bio-Inspired Artificial Intelligence. Completed courses in Computer Vision, Digital Signal Processing, and Advanced Computer Networks, building a foundation for my transition into machine learning research.

Earned a Bachelor of Science in Electrical Engineering with a specialization in Electronics. Designed an LED lamp driver as part of my final project and developed an Android application to control the lamp’s brightness. Also served as a teaching assistant for multiple electronics courses, strengthening my foundation in hardware and embedded systems.