Is EigenPal profitable in 2026?

No revenue has been disclosed. EigenPal has raised only a $500K Y Combinator convertible note that closed around January 27, 2026, and is at pilot stage with 2 European banks, not confirmed paying contracts.

Could AI replace EigenPal's own product?

Possibly. Anthropic's Claude gained the ability to learn workflows by watching screens and OpenAI launched a ChatGPT Work agent around July 26, 2026, both threatening to commoditize the document-workflow inference EigenPal sells.

Who founded EigenPal?

EigenPal was founded in 2025 by Matej Novak and Jedrzej Blaszyk, according to a profile published January 29, 2026. The company joined Y Combinator's Winter 2026 batch shortly after.

Is Rossum a bigger threat than EigenPal?

Yes by scale. Rossum is described as serving 450+ enterprise customers as of May 26, 2026, per competing vendor Kognitos, versus EigenPal's vendor-claimed 2 European bank pilots as of January 26, 2026.

EigenPal is privately held by its founders, Matej Novak and Jedrzej Blaszyk. Its only disclosed outside capital is a $500K Y Combinator convertible note that closed around January 27, 2026; no other investors are disclosed.

Does EigenPal have real paying customers?

EigenPal (vendor-claimed) says it is working with 2 large European banks on loan automations as of January 26, 2026, but no independent press had confirmed named enterprise logos or paid contracts as of May 27, 2026.

YC Winter 2026 · United States

EigenPal

Early-stage AI platform that extracts and automates document workflows for regulated enterprises.

Protection under test

Assessed 27 July 2026 · Evidence as of 27 July 2026

Something worth protecting exists and is being tested right now.

On the recordSince 27 July 2026 · Checkable from mid-2027 · Every call on the record Checkpoint: Whether the two claimed EU bank pilots convert into renewed, paid on-prem production contracts

The call

EigenPal has pilots and a compliance pitch but no confirmed paid contracts, sitting between commoditizing AI extraction and better-funded document-AI rivals.

EigenPal helps banks and regulated companies pull information out of documents automatically, with a focus on keeping data on the company's own computers for security and legal reasons. The problem is that the basic extraction technology is getting easier to copy as big AI models like Claude and ChatGPT get smarter at learning workflows just by watching what employees do, and bigger rivals like Rossum and Hyperscience already have hundreds of enterprise customers and deeper resources. The one thing that could save EigenPal is if their two claimed pilot projects with large European banks actually turn into real paid contracts where customers stick around because they trust EigenPal's setup and don't want to switch, but this hasn't been proven yet and those pilots are still just tests.

Why

AI pressure: Anthropic Claude workflow-watching agent and OpenAI ChatGPT Work. Frontier models learning to infer document workflows from a few examples could absorb EigenPal's core extraction layer. Could AI do the core job? Yes, likely.

Bigger players: Rossum, Hyperscience, ABBYY, and hyperscaler document APIs (Textract, Google Document AI, Azure Document Intelligence). Rossum alone serves 450+ enterprise customers with on-prem European deployments already in the same regulated-finance space EigenPal targets. Could a giant copy it? Yes, easily.

What protects it. Claimed lock-in from on-prem deployment and compliance trust with two EU banks, still unproven past the pilot stage.

What would change this call

Whether the 2 claimed EU bank pilots convert to paid, renewed on-prem contracts by mid-2027. Would show workflow lock-in and compliance trust are real, not just pilot activity.
Anthropic or OpenAI shipping a general workflow-inference agent that reaches regulated document processing. Would show the core extraction task is being commoditized by frontier models.
Whether EigenPal's marketing narrows from 'AI Document Workflows for Enterprises' to a specific regulated corridor. Would show EigenPal is executing the niche-wedge strategy needed to survive.

Key numbers

$500K raised via YC convertible note (2026-01-26)
2 large European bank pilots claimed (2026-01-26)
Deployable on AWS, Azure, GCP, or on-prem (2026-05-15)

Named risks

Core extraction and workflow-inference commoditized by frontier models (Claude, ChatGPT Work) as of mid-2026. Severity: high
Only 2 pilot relationships with European banks; no confirmed conversion to paid production contracts. Severity: high
Crowded IDP market with better-funded specialists (Rossum, Hyperscience, ABBYY) offering similar eval/audit features. Severity: medium
Generic 'AI Document Workflows for Enterprises' positioning lacks a defended vertical wedge. Severity: medium
Conflicting founder/HQ location reporting (San Francisco vs Warsaw-and-London) signals thin public information. Severity: low

Rivals, as read in July 2026

Rival	Direction	Evidence
Rossum	losing to	450+ enterprise customers claimed vs EigenPal's 2 bank pilots, as of 2026-05-26
Hyperscience	contesting	Listed as closest IDP peer competing for same regulated-enterprise buyers, 2026-05-04
ABBYY	contesting	Legacy IDP leader named as adjacent incumbent in document-extraction market, 2026-03-29
Amazon Textract, Google Document AI, Azure Document Intelligence	losing to	Hyperscaler APIs commoditizing core extraction layer EigenPal depends on
Anthropic Claude / OpenAI ChatGPT Work	losing to	Frontier agentic workflow-learning features shipped around 2026-07-26, same core value prop

Questions people ask

Is EigenPal profitable in 2026?: No revenue has been disclosed. EigenPal has raised only a $500K Y Combinator convertible note that closed around January 27, 2026, and is at pilot stage with 2 European banks, not confirmed paying contracts.
Could AI replace EigenPal's own product?: Possibly. Anthropic's Claude gained the ability to learn workflows by watching screens and OpenAI launched a ChatGPT Work agent around July 26, 2026, both threatening to commoditize the document-workflow inference EigenPal sells.
Who founded EigenPal?: EigenPal was founded in 2025 by Matej Novak and Jedrzej Blaszyk, according to a profile published January 29, 2026. The company joined Y Combinator's Winter 2026 batch shortly after.
Is Rossum a bigger threat than EigenPal?: Yes by scale. Rossum is described as serving 450+ enterprise customers as of May 26, 2026, per competing vendor Kognitos, versus EigenPal's vendor-claimed 2 European bank pilots as of January 26, 2026.
Who owns EigenPal?: EigenPal is privately held by its founders, Matej Novak and Jedrzej Blaszyk. Its only disclosed outside capital is a $500K Y Combinator convertible note that closed around January 27, 2026; no other investors are disclosed.
Does EigenPal have real paying customers?: EigenPal (vendor-claimed) says it is working with 2 large European banks on loan automations as of January 26, 2026, but no independent press had confirmed named enterprise logos or paid contracts as of May 27, 2026.

Timeline

2026-07-26Anthropic Claude gains screen-watching workflow learning; OpenAI ships ChatGPT Work agent
2026-05-27LlamaIndex notes no independent press confirmation of EigenPal enterprise logos
2026-05-26Competing vendor Kognitos cites Rossum at 450+ enterprise customers
2026-05-15Claude's Corner profiles EigenPal as eval-first platform; HQ location unresolved
2026-05-04Docsumo lists EigenPal peers: Rossum, Reducto, Extend, Nanonets, Hyperscience
2026-03-29Klearstack lists incumbent hyperscaler document-AI APIs and ABBYY as adjacent giants
2026-01-29Profile confirms EigenPal founded 2025 by Matej Novak and Jedrzej Blaszyk
2026-01-27EigenPal closes $500K Y Combinator convertible note
2026-01-26EigenPal launches publicly on YC Launch as part of W26 batch

Revision history

27 July 2026

First assessed. Protection under test. The text above is fixed as of this date and is not edited afterwards.

In the same band

Sequence Markets Piris Labs Inviscid AI