claudiosv / codebert

CodeBERT

👋   Hey there, I'm Claudio Spiess

I'm a data scientist and software engineer. I'm currently pursuing a Master's in Computer Science at the University of California, Davis. My work revolves arround applying machine learning techniques, especially from the natural language processing community, to software engineering problems.

🧠   Expertise: Machine Learning, Data Science, Python, ML4SE, AI4Code

🔗  [Personal website] - [LinkedIn]

🎓   Education

• Master of Science in Computer Science at University of California, Davis, USA (In Progress)
• Bachelor in Computer Science & Engineering at Free University of Bozen-Bolzano, Bolzano, Italy (2019)
• Computer Science Study Abroad at College of Charleston, Charleston, USA (2017)

👨‍💻   Experience

• Graduate Student Researcher at University of California, Davis

  • Perform research on machine learning for software engineer problems, currently investigating use of BERT embeddings of trace data for code retrieval tasks. Extensive use of data analysis, deep learning, and NLP techniques.

• Teaching Assistant at University of California, Davis

  • ECS 160 Software Engineering: Held discussion section & office hours, produced & graded homework, proctored & graded exam, answered student questions on discussion forum

• Machine Learning Intern at Overstock.com

  • Built & deployed Streamlit app to analyze embedding spaces used in product recommender systems
  • Designed & implemented 3 core features: nearest neighbor visualizations, linear & classification probes, and catalog coverage analysis for recommender systems ML team, optimizing their development and debugging workflow

• Junior Software Engineer at Floryn

  • Developed features & improvements on Ruby on Rails and React/TypeScript apps that managed over C185 million euros in outgoing loans at one of the Netherland’s fastest growing startups with over C78 million in funding
  • Worked within Data Science team to deploy & maintain Dockerized production Python ML inference services
  • Created system for monitoring ML model outputs & explaining decisions to non-technical stakeholders e.g. through the use of SHAP

• Full Stack Development Intern at Datatellers Srl

  • Wrote a web app to manage geospatial data for a large engineering company using Bootstrap, jQuery, and PostgreSQL

🛠️   Skills

• Languages:   Python, Ruby, SQL (Postgres), TypeScript
• Frameworks:    PyTorch, Flask/FastAPI, Rails, React, Node.js
• Developer Tools:       Git, Docker, Terraform, AWS, CircleCI, VS Code
• Libraries:       pandas, NumPy, Huggingface, Matplotlib, plotly