I clean data, run EDA, engineer features, and build machine learning models and dashboards that answer business questions and support decisions.
Created an AI system that uses Twitter sentiment and topics to retrieve financial news. Used RAG with LLMs to build a smart query engine.
Stack: Python, Transformers, Hugging Face, OpenAI, ChromaDB
PublicationEngineered features, visualized trends, and built predictive models to forecast market shifts. Added Reddit sentiment analysis to support insights.
Stack: Python, Pandas, Plotly, Reddit API
GitHubBuilt a real-time scraping pipeline to collect and structure Instagram post data. Enabled internal teams to analyze marketing signals.
Stack: Python, Selenium, MySQL
Developed a data pipeline to parse social media and forecast public sentiment. Detected early sentiment shifts before trends spiked.
Stack: PySpark, MongoDB, Dash, NLTK
GitHub“Kagan was fully committed during his time at Orcawise. He gained hands-on experience in data engineering, visualisation, NLP, fine-tuning, and prompt engineering. He’s a strong communicator and supported peers at AI networking events. We’re happy to recommend him.”
– Kevin Neary, CEO at Orcawise
“Kagan worked with me while I mentored him at Orcawise. He’s hardworking, asks smart questions, and consistently delivers. He has strong skills in data visualisation, NLP, LLMs, data scraping, and SQL. He’d be a valuable asset to any team.”
– Sanpreet Singh, Head of AI at Orcawise