Show HN: Kairos – Open-source risk and fraud decisioning platform

15 days ago (github.com)

Hey HN, Ayan here from Finic. We’re launching Kairos, an open source alternative to Sift, Sardine, and other real-time fraud decisioning platforms. Kairos addresses two of the biggest problems facing traditional decisioning tools: a lack of labeled data to train on and dealing with fraudsters that are constantly changing their tactics.

1. Traditional fraud solutions use ML models such as XGBoost to detect fraud, but these models are dependent on quality labeled data which many businesses do not have. We address this by using “System One” decision models in additional to traditional models such as XGBoost and logistic regression. This allows you to incorporate plain language questions such as “Is this a throwaway email” into your decisioning, bypassing the need for large amounts of labeled data. This also allows you to capture aspects of fraud that are more unstructured, which is something that XGBoost models struggle with.

2. Fraud teams are facing off with live players that are constantly changing their behavior to bypass the detection logic they have in place. You can train a detection model or define a ruleset once, but it will become obsolete in a matter of weeks as fraudsters figure out what your detection logic is. This means that you need a team regularly training models and updating rules based on new fraud patterns. Typically this takes a large data science/ML team. Kairos addresses this by introducing an MCP server, so that your agent (Claude Code, Cursor, Codex, etc.) can define a decision workflow, train models on your data, and pick the variant that performs best.

You can self-host Kairos with Docker or use the hosted version at https://kairos.finic.ai

I love that this is open source, thank you. I'm not a fan of TypeScript on Node for this solution space. For a payment gateway, I'd want a program that can't modify itself from external sources. Programs in this runtime can rewrite themselves while running, and the Dockerfile doesn't lock it down. So as it stands, nothing at the runtime level prevents prototype pollution. And a config defect is the kind of bug that keeps on giving.

This isn’t really a new problem. Back in 1995, I was the Sr. Internet Mail Administrator at AOL. We were doing everything we could to fight the spammers.

We found that we would work on a new technique over a period of weeks or months, and then they would find a way to work around our technique in a single weekend.

Rinse and repeat.

The only thing that has changed is that the market for scammers has moved, and they’ve gotten a lot better and faster at adapting to your protection techniques.

Not something I need but super cool that this is both open source and shows the realtime flow.

Thanks for showing Keiros! This looks very promising!

Could you explain how "System One" differs from the traditional, rule based, approach? Sticking to your example: How would the question 'Is this a throwaway email?' be translated into Kairos business logic?

When using the MCP Server: How does Claude develop such models, e.g. feature engineering? Is this an MLOps pipeline behind an MCP?

Why would you ask a decision model if an email is disposable? There is so much more behind that than the looks of the email. This just seems like yet another video-coded anti-fraud thingi.

Is anybody else annoyed by made-up words like "decisioning" and "proofing"?

You can't slap "ing" on a noun. It's for verbs. Stop nouning.

  • Language is what people use to communicate, not what's written in the books. It's fluid, and adapts to requirements as the requirements change.

    Would it be better if they invented a completely new word, with no clue as to what it means, rather than adapt an existing language structure to their purpose, which everyone understands immediately (even if they get grumpy at it)?

There's a meta linux distribution with that name already :(

https://kairos.io/