Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data

news.ycombinator.com

Post reply on HN
Print view View original

101–110 of 124 posts

Re: Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data

#102

Post by harryf »

makk wrote: post #93

> a major commercial bank I work with couldn’t improve credit risk models because critical data was stuck in PDFs and emails. Great use case! Worked on exactly this a decade ago. It was Hard™ then. Could only make so much progress. Getting this right is a huge value unlock. Congrats!

Make sure you have an on-premise option for this type of customer. I've worked at two software companies in Europe with tangentially similar products related to document analysis. On premise is a key requirement.

Even though it's 2024, banks, financial institutions like insurance companies etc. tend to be _very_ cautious with valuable documents involving customers. There are also regional regulations that prevent things like patient data being shared with _any_ 3rd parties. Even one of the big 4 oil companies that I've dealt with as prospective customer - very strict rules requiring on premise solutions.

The good news is many are using things like Kubernetes and OpenShift internally, so it should be possible to port what you do on AWS to on-premise.

Re: Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data

#105

Post by PeterStuer »

wjnc wrote: post #65
mritchie712 wrote: post #20

Earlier quoted context omitted.

> 80% of enterprise data is unstructured I've seen quotes like this many times. It's silly. I worked at a big bank for over a decade. 95% of the data we cared about was already in a SQL database. Maybe ~80% of our data was "unstructured", but it wasn't stuff we cared about for risk management or other critical functions. > people are not willing to pay more money for less risk I'd disagree here. Banks are willing to…

Who, in your example, put the customers financial data in the SQL database? Because in my part of finance that’s either the customer, or an employee. Our customers are asking for integration with a lot of their systems (say HR / patrolling), but never ever offer to hook up their accounting system. If we want financial data, we either get a PDF with their audited financial statement or in exceptional cases a custom au…

I created automated descision suppport systems in asset based finance. For daily needs you get customer financials and other risk data from both official national sources and the likes of Dunn and Bradstreet, Graydon etc. The choice of providers depends on both the customer and deal risk/size. While the "api"'s to these providers might be clunky (putting structured request file on an ftp server and polling for a response), the data is structured (enough) to process.

Deals that are exceptional enough get assigned to a risk officer that deals with it as a case (sidenote: they use a lott of selfmade Excel, VB and low-code tools as they never get IT priority for these cases). There is not enough uniformity as well as a decreased tolerance for inaccuracy in those to warrant extensive automation.

Re: Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data

#106

Post by wjnc »

PeterStuer wrote: post #105
wjnc wrote: post #65

Earlier quoted context omitted.

Who, in your example, put the customers financial data in the SQL database? Because in my part of finance that’s either the customer, or an employee. Our customers are asking for integration with a lot of their systems (say HR / patrolling), but never ever offer to hook up their accounting system. If we want financial data, we either get a PDF with their audited financial statement or in exceptional cases a custom au…

I created automated descision suppport systems in asset based finance. For daily needs you get customer financials and other risk data from both official national sources and the likes of Dunn and Bradstreet, Graydon etc. The choice of providers depends on both the customer and deal risk/size. While the "api"'s to these providers might be clunky (putting structured request file on an ftp server and polling for a resp…

Thanks. I see the context now. Our asset managers are indeed lucky to have Bloomberg and such, which are easily integratable (and indeed, have been "SQL" for more than a decade now). I'm aware of the third party providers of customer financial information. Lucky to operate in a niche that is not served by them. Graydon (the only one I've been in contact with) is facing a massive disruption though. Their higher tiers are perhaps not enterprise expensive, but Trellis and the likes are probably more integratable and more affordable.

But still, the one building the LLM-integration is 4x as expensive as the one manually entering the data. It's all about TOC, scale and risk perception. I also love that "risk people" (in the banking context, I'd say: model people) think their data quality should be exceptional and then use end user computing MacGyver style models. Spit and popsicle sticks.

Re: Launch HN: Trellis (YC W24) – AI-powered workflows for unstructured data

#108

Post by ace32229 »

makk wrote: post #93

> a major commercial bank I work with couldn’t improve credit risk models because critical data was stuck in PDFs and emails. Great use case! Worked on exactly this a decade ago. It was Hard™ then. Could only make so much progress. Getting this right is a huge value unlock. Congrats!

I have just been working through the same problem (though just PDFs). Google DocAI helped enormously after a bit of initial input.
Post reply on HN