# mckaywrigley/paul-graham-gpt

RAG on Paul Graham's essays.

Repository: https://github.com/mckaywrigley/paul-graham-gpt
Canonical: https://ross.abutalabs.com/products/paul-graham-gpt
Homepage: https://JoinTakeoff.com
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2023-07-28T17:50:55+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 91
- inputs: {"age_days": 1281, "days_push": 1132, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2663, forks 372 (observed 2026-08-28T04:07:08.928124+00:00)

## What it is
An open-source AI-powered search and chat application built over Paul Graham's essays using OpenAI embeddings, pgvector on Supabase, and GPT-3.5-turbo. It includes scraping and embedding scripts plus a downloadable CSV dataset of text and embeddings.

## Use cases
- ask questions about Paul Graham's essays
- semantic search over a corpus of essays
- build a RAG chatbot over scraped web content
- learn how to embed and store text with pgvector
- chat with a personal document collection

## When to avoid
- you need production-grade RAG with advanced retrieval, evaluation, or multi-tenant support
- you want to avoid OpenAI API costs or vendor lock-in
- you need a general-purpose document ingestion pipeline beyond a single scraped site

## Facets
- artifact type: application
- maturity: maintenance
- function: rag, search-engine, web-scraping, llm-inference
- domain: large-language-models, education
- platform: self-hosted
- tags: paul-graham, openai-embeddings, pgvector, supabase, nextjs, semantic-search, chatbot, essays, retrieval-augmented-generation, natural-language-processing, nodejs, web-server

## Member repositories
- mckaywrigley/paul-graham-gpt (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:08.928124+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:18:20.828029+00:00, confidence not recorded.
  - readme: https://github.com/mckaywrigley/paul-graham-gpt (fetched 2026-08-28T04:07:08.928124+00:00, sha ac57929a5b62)
  - homepage: https://JoinTakeoff.com (fetched 2026-08-29T10:01:26.423867+00:00, sha 29e9f961b06f)
  - site_page: https://www.jointakeoff.com/about (fetched 2026-08-29T10:01:26.430178+00:00, sha 2b10fb196859)
  - site_page: https://www.jointakeoff.com/pricing (fetched 2026-08-29T10:01:26.426380+00:00, sha f96920d1f316)
  - site_page: https://www.jointakeoff.com/changelog (fetched 2026-08-29T10:01:26.428250+00:00, sha 018395019c4f)
- Data as of 2026-08-30T08:39:29.467469+00:00.