Irsa Ashraf
Data Scientist + Backend Engineer
Work: Turning messy data into insights via APIs and ML.
Off-the-clock: Biking, NYC Footy, and staying on top of fashion trends.
Data Scientist + Backend Engineer
Work: Turning messy data into insights via APIs and ML.
Off-the-clock: Biking, NYC Footy, and staying on top of fashion trends.
I'm a Data Scientist and Backend Engineer working at the intersection of data science and enterprise scale analytics. I've built production-level microservices and APIs that process and present complex data at scale, and ML models that deliver actionable insights.
As a self-taught programmer, I love building my technical skills through hands-on projects and real-world problem-solving.
Outside of work, I'm biking around NYC, playing soccer with my NYC footy team (Go FC Chaos!), trying to get into the 8am Solidcore class, and writing about data and tech on Substack.
LLM-powered agent using OpenAI's API to interpret natural language queries about NYC tennis courts. Implements agentic RAG with tool-calling architecture, geocoding, and real-time distance calculations via Dockerized FastAPI.
Publishing insights on data science, machine learning, and technology on Substack. Sharing practical experiences, technical tutorials, and thoughts on the evolving landscape of AI and analytics.
Data analysis project exploring eating habits and restaurant trends across New York City neighborhoods. Analyzes patterns in cuisine types, pricing, ratings, and geographic distribution using Python data science tools.
Multi-class text classification model identifying toxic comments from Wikipedia and YouTube. Preprocessed and balanced datasets, implemented NLP techniques including Bag of Words and CNN architectures for classification.
Unsupervised learning project clustering Spotify songs into genres based on audio features. Applied dimensionality reduction and clustering algorithms to discover patterns in music characteristics and genre boundaries.
I'm always open to discussing new projects, opportunities, or just having a conversation about data, ML, and technology.