Scholar
Nick McKenna
Google Scholar ID: 9MzRL9YAAAAJ
GitHub Applied Science
Citations & Impact
Citations
353
H-index
6
i10-index
5
Publications
10
Co-authors
11
list available
Publications
Resume (English only)
Background
- Applied researcher currently working on GitHub Copilot, focusing on model contextualization to enhance model performance in static and agentic modes through smart context.
- Broadly interested in language model capabilities in textual understanding, coding, and tool use, as well as mitigating LLM hallucinations.
- Pursues models that leverage external, human-editable data sources such as structured knowledge and open-domain text, using tools and code to accomplish tasks precisely.
- In code generation, has investigated approaches to improve program synthesis in low-resource programming languages, LLM dialogues verified by code generation and execution, and complex data science task completion.
- In NLP, has extensive experience in natural language inference (NLI), particularly for question answering, summarization, and Knowledge Graph usage via lexical semantic modeling.
