Skip navigation
Tutorial: Automating KYC with AI agents 🪪🕵
0:00 / 1:18

Tutorial: Automating KYC with AI agents 🪪🕵

Jerry Liu

@jerryjliu084K followers

Visit on X
103 likesDownloadOriginal post

30,062 views

Tutorial: Automating KYC with AI agents 🪪🕵 I’m creating a new tutorial series of automating practical document workflows with agents. Every financial institution needs to perform KYC (know your customer) to verify a customer’s identity, and this involves manually sifting through IDs, bank statements, etc. and doing the cross-checking by hand. This is a great first use case for agentic document workflows: 1. Extract identification information from the user supplied ID (license, passport) 2. Extract fields from utility bills/bank statements and then use LLMs to cross-validate extracted fields with the extracted ID fields It obviously doesn’t cover the full e2e process and uses publicly available online data, but should be a good reference guide to get started. To make this work well, you do need high-quality document extraction with confidence scores and citations! Check out the tutorial: If you’re interested come check out LlamaParse:

8 Comments

Comments from the original post on X (Twitter)

  • Sahrim Alija's profile picture

    @venditorix

    @llama_index Great Jerry. That's exactly what I need. Do you have any other tutorials on how to create a rag for insurance purposes? Can you give me some advice for my project? Is it possible to contact you?

  • hira's profile picture

    @Hiraweb3

    love seeing practical agent use cases like this

  • Max Slinger's profile picture

    @PromptSlinger

    automating the part where someone squints at a passport photo and goes 'yeah thats probably the same person' was always inevitable tbh

  • MacGuffin's profile picture

    @MacGuffin_Tech

    Great starter use case for doc‑centric agents. KYC is basically the perfect playground: extraction + cross‑validation. The real differentiator will be extraction quality (scores + citations). Can’t wait to try the LlamaParse tutorial.

  • Becker Meister's profile picture

    @MeisterCoins

    zamn pydantic schemas making document parsing reliable in 2024? we might actually solve kyc automation fr

  • Massively Parallel Procrastinator's profile picture

    @SHELLEYBLEND

    Thank you again, found this when I came to get the liteparse_samples!

  • TLAY's profile picture

    @tlay_io

    Agentic KYC needs evidence boundaries. Extracted fields should carry source citations and confidence, conflicting records should escalate, and sensitive documents should stay minimized. Automation is valuable when every verification decision remains reviewable.

  • Myke86d's profile picture

    @myke86d

    Love it thank you for sharing did the same but in