We won the Open Agentic AI Innovation Award at the first AAIF Technical Excellence Awards.We won the Open Agentic AI Innovation Award Read why

AI invoice processing from inbox to ERP posting

Invoices matched to the order before they reach AP

An agent reads each invoice, matches it to the purchase order and receipt, and hands your AP team the exceptions and the approvals you choose.

Book a free consultation How we build agents
ap-invoices

  1. Intake agentAgent

    1. inbox.fetch_invoice supplier email and PDF
    2. parser.read_invoice header, lines and tax

    A supplier invoice is read from the AP inbox, scanned pages included.

  2. Matching agentAgent

    1. erp.get_purchase_order order lines and prices
    2. erp.get_receipt goods received
    3. match.three_way invoice, order, receipt in code

    Quantities match, but one unit price is above the purchase order. Approve the difference or send it back?

    Needs a personPrice above the purchase orderOne line differs from the agreed price
    Approved by AP lead

  3. Posting agentAgent

    1. erp.post_invoice coded to the cost centre

    Invoice posted with the match result and the approval, ready for the next payment run.

Example data. An agent at work in a team channel, with a person approving the step you choose.
Mixam: conversion to a final sale from about 20% to about 40%
Where agents help

Invoices wait in the inbox while AP retypes them.

Agents do the reading and matching, and your AP team approves what gets paid.

  1. 01

    Where it stallsAP staff retype the same fields from every invoice.

    What we do An agent reads supplier, dates, lines and tax from PDFs and scans, each pinned to its place on the page.

  2. 02

    Where it stallsMismatches with orders surface at payment time.

    What we do Code compares each invoice with the purchase order and the receipt before anything is posted.

  3. 03

    Where it stallsEvery supplier sends a different layout.

    What we do The agent reads fields by meaning across layouts. A field it is unsure of goes to a person.

What we build

What we build for invoice processing

Whatever comes in, every project is built from the same parts below, set up for your files and your systems.

Invoices we process

  • Supplier invoices
  • Credit notes
  • Purchase-order matching
  • Utility and service bills
  • Foreign-currency invoices
  • Scanned and PDF invoices
  • E-invoices
  • Expense receipts

How we build it

  • Invoice reading and extraction

    Supplier, dates, line items and tax read from PDFs and scans, each field linked to its source.

  • Order and receipt matching in code

    Invoice lines checked against the purchase order and goods receipt by rules your team can test.

  • ERP and inbox connections

    Coded invoices posted to your ERP through its API, with the access rights you set.

  • Approval rules your AP team agrees

    Tolerances and stop rules agreed with your finance team and enforced in the system.

Not seeing your case? Bring the invoices that need the most manual work and we start from those.

Tell us about your workflow
How we build agents

Agents that work as part of your team

They work in the tools your team already uses, take on whole tasks and stop for a person where you decide.

  1. Routines: agents that start on a schedule, on new GitHub issues or on an API delivery.
  2. Connected systems: GitHub, Linear, Notion, Slack, Jira and other tools an agent may call, each connected with its own sign-in.
  3. Approvals log: each action that waited for a person, with who asked, who decided and when.
Routines: agents that start on a schedule, on new GitHub issues or on an API delivery.
Connected systems: GitHub, Linear, Notion, Slack, Jira and other tools an agent may call, each connected with its own sign-in.
Approvals log: each action that waited for a person, with who asked, who decided and when.

Screens from the workspace our agents run in, with example data.

Where it runs

Teams and industries that use it

Each page shows where agents fit in that team or industry.

Engineering in the open

We help build what agentic AI runs on

Open standards, open-source tooling and published research, all public, so you can check our engineering before you hire us.

Shaping agentic AI

Agentic AI Foundation

Vstorm was the first agentic AI engineering company to join the Agentic AI Foundation and received its 2026 Open Agentic AI Innovation Award.

Read the announcement

Open source

We build open-source agent tooling that other engineering teams install

10M+ PyPI downloads
4,000+ GitHub stars
33 public repositories

Framework contributors

When the framework breaks, we fix it at the source

We maintain our own open-source agent framework and contribute fixes to the frameworks our agents run on, through official partnerships with their teams. Breaking changes reach us before they reach your codebase.

Research

Our engineers and PhDs publish peer-reviewed research

Peer-reviewed papers on agent architectures, evaluation and AI security, and an open benchmark anyone can rerun against our results.

Track record

Frontier AI teams now build on what we shipped first

Our client agents were in production before most of today's agent frameworks had a stable release. You get a team that already knows what breaks between a promising demo and a system people rely on every day.

Case studies
All case studies
Engagement models

Ways to start

Each path starts with a free call with a founder or senior AI engineer. A build goes ahead only after its Proof of Value meets the agreed threshold.

Start with one workflow

We map one workflow with your experts and agree a success threshold before any build.

Best whenYou want the success test agreed before you commit to a build.

Book a free consultation
  1. Process redesign and build Best whenYou want one team accountable for the outcome.
FAQ

AI invoice processing, answered

What is AI invoice processing? +
AI invoice processing uses an agent to read supplier invoices, match them to purchase orders and goods receipts and post them to your ERP. The agent reads PDFs, scans and email bodies, pins each field to its place on the page and hands mismatches to your AP team. Matching and posting rules run in code, and a person approves the payments you choose.
Which invoices should accounts payable automation start with? +
Invoices that carry a purchase order and arrive by email, from suppliers who send many of them. The order and the receipt give the agent something to match against, so errors show up early. Measure field accuracy on a few weeks of your past invoices before going live, then add invoices without orders and new supplier layouts.
Have you built AI invoice processing before? +
We have no published case on invoice processing yet. The closest builds read and check documents the same way. For a US health insurer, claim processing was cut from 3 hours to 8 minutes, with a specialist still deciding. For a real estate client, due diligence was cut from weeks to minutes. For a US telecom, 98% of device activation workflows were automated.
How does the agent handle scanned invoices and new layouts? +
Each field is pinned to its place on the page, so a reviewer can check it against the scan. When the agent is unsure about a field, or the invoice does not match the order, it stops and asks a person. Corrections from your team go into the test set, so every new version is checked against the layouts that caused trouble before.
How do you know it works before the full build? +
We agree a success threshold with your team before the build and test the agent on your own cases in a Proof of Value. It measures accuracy and running costs, and the build goes ahead only when the threshold is met. Corrections from your team go back into the test set, so every later version is checked against them.
Who owns the system, and which models does it run on? +
You own the code, prompts, evaluation sets and rules by contract. The agents run in your environment on hosted or open-weight models. The model sits behind an interface in code, so you can change it when accuracy, speed or your hosting rules change, without rebuilding the agent.
Start with one workflow

Test one workflow on your own data

A Proof of Value shows accuracy and running costs before the full build.

Book a free consultationRead case studies
From the blog

Reading on document extraction

All articles
Engineering

When clean text is not enough: structured extraction for RAG

When Clean Text Is Not Enough: Structured Extraction for RAG enables richer retrieval and accuracy—explore smarter solutions.
Engineering

Beyond Frontier Models: Testing Lightweight LLMs for Document Processing in RAG

Why bother with small models when capable LLMs are available? Because businesses may face stricter requirements than inidividual users
Engineering

What is Information Extraction using LLMs?

IE emerges in making sense of this data deluge, transforming unstructured data into structured actionable information. AI for companies