LLM usage log

Brainstorming

Idea Generation

LLM tools were instrumental in brainstorming the initial ideas for the project. They provided multiple perspectives, allowing us to explore different approaches to solving key challenges.

Refining the Project Plan

LLM tools offered constructive feedback on the proposed workflows, helping us streamline the project structure and prioritize tasks.

Feature Engineering Suggestions

For machine learning tasks, LLMs were used to suggest meaningful features to include (e.g., review length, word frequencies) and methods to preprocess textual data effectively.


Writing

Reformatting Text

  • Transforming Lists into Prose
    LLM tools helped convert raw, bullet-pointed notes into structured and well-written prose, improving clarity and readability.

  • Report Structure Improvements
    The tools suggested better organization of project reports, ensuring smooth flow and logical connections between sections.

Proofreading and Editing

  • Grammar and Style Checks
    LLM tools were used to proofread technical documents, catching grammar errors, awkward phrasing, and inconsistencies in tone.

  • Technical Precision
    For project-specific content, such as methodology explanations, LLM tools enhanced the technical accuracy while keeping the language clear and concise.

Summarization

  • Literature Reviews
    LLM tools summarized lengthy academic papers and documentation, extracting key points relevant to the project. This saved significant time during the literature review process.

  • PDF Summarization
    LLMs processed technical PDFs (e.g., research papers, documentation) and provided concise summaries, enabling quick comprehension of complex material.


Code

Code Commenting and Documentation

  • Adding Explanations
    LLM tools were used to automatically generate comments for existing code, explaining key functions, logic, and purpose. This improved readability and maintainability for team collaboration.

  • Explanatory Documentation
    The tools produced comprehensive documentation for code components, including class and function-level descriptions.

Code Optimization Suggestions

  • Performance Improvements
    LLM tools suggested optimizations for repetitive or inefficient code, such as reducing redundant computations or improving loops.

Debugging and Error Resolution

LLM tools were used to debug errors efficiently by analyzing tracebacks and suggesting corrections. This accelerated the troubleshooting process and minimized downtime.

Code Refactoring

  • Readability Enhancements
    LLMs helped refactor complex, lengthy code into modular and cleaner components.

  • Best Practices
    The tools ensured adherence to Python best practices (e.g., PEP 8 standards).


Visualization and Optimization

Data Visualization

  • Chart Recommendations
    LLM tools suggested appropriate visualization techniques (e.g., line charts, scatter plots, word clouds) based on the nature of the data.

  • Labeling and Styling
    Enhanced visualizations with better axis labels, titles, and legends for clarity and professional presentation.