Skip to main content
Alph uses AI-powered search that understands meaning, not just keywords. How it works:
  1. Notebooks are converted to vector embeddings
  2. Your query is embedded and compared
  3. Results ranked by semantic similarity
Example queries:
  • “sentiment analysis with transformers”
  • “customer churn prediction”
  • “time series forecasting LSTM”
Natural language queries work better than keywords.

Notebook-Level

Search by title, description, and overall purpose

Cell-Level

Search within individual code cells for specific implementations
  1. Go to Notebooks (global, not within an org)
  2. Enter your query in natural language
  3. Filter by tags, author, or date
  4. Sort by relevance, trending, or recent

​
Browse by Tags

Explore notebooks by topic:
  • Languages: python, r, julia
  • Domains: machine-learning, data-viz, nlp
  • Level: beginner, intermediate, advanced

​
Improving Your Notebooks’ Discoverability

Descriptive titles: “Predicting Housing Prices with Random Forest” not “Notebook 1” Informative descriptions: What it does, what techniques are used, what readers will learn. Relevant tags: Use specific, searchable keywords (max 10). Documentation: Well-documented notebooks with markdown rank higher.