Skip to content

Welcome to the

0

Labs experiment

Research crew: a team of agents that researches a market and writes the brief

A team of agents that researches a market question and writes the brief. A lead agent splits the question, a searcher collects sources, a reader summarises each one, an analyst compares claims and flags conflicts, and a writer delivers a two-page brief with its sources.

Where this pattern fits

  • Founders and analysts who need a first market scan quickly
  • Competitive research that repeats every quarter
  • Any research task where sources must be kept and checked
  • LangGraph
  • Claude
  • Web search
  • Python

How it works

Research questionLead agentSearcherReaderAnalystWriterWeb searchScraperNotes store

Select any node to see what it does.

Simulated run
  1. 00.00leadsplit the question into 5 sub-questions
  2. 01.20searcher38 sources collected
  3. 03.05analyst3 competing claims flagged
  4. 04.40writertwo-page brief with sources ready

The architecture, step by step

Each part of the diagram, in the order a run moves through it.

  1. Research questionWhere every run starts.
  2. Lead agentSplits the question and assigns the research.
  3. SearcherFinds sources on the web.
  4. ReaderReads and summarises each source.
  5. AnalystCompares claims and flags conflicts.
  6. WriterWrites the final brief with sources.
  7. ToolsThe agents act through the web search, scraper and notes store.

How it is built

The design choices that make the pattern hold up with real users.

visitor@azan-portfolio: ~

_ ___ ___ _____

/_\ ____ __ _ _ _ / __| _ \_ _|

/ _ \ |_ / / _` || ' \ | (_ | _/ | |

/_/ \_\/__| \__,_||_||_| \___|_| |_|

Hi, I am AzanGPT. Ask me about Azan's work, stack, services or how to reach him.

Type 'help' for commands, or ask a question. Tab completes, the arrow keys walk your history.

AzanGPT runs on an AI model that knows this site. It can get things wrong, so check anything important with Azan.