We built Fieldiq because 5% error rates are not a cost of doing business.
Carlos Mendes ran ops automation projects at a Houston logistics company where the BPO contract had a 4.8% error rate SLA — and was running above 7% in practice. Every vendor demo worked. Production accuracy didn't hold. Fieldiq is the system he kept needing and couldn't buy.
What Carlos saw that we couldn't unsee
In 2022, Carlos Mendes was running ops automation projects at a mid-market logistics company in Houston. The company was processing 40,000 invoices and freight bills per month through a BPO at $21 per document. The BPO contract had a 4.8% error rate in the SLA. In practice the actual error rate was running above 7%.
Every evaluation of document AI vendors ended the same way: the demos worked, the production accuracy didn't hold, and exception handling meant sending documents back to the same BPO floor for rework. There was no system that actually owned the exception outcome.
Carlos founded Fieldiq in 2023 with a specific design requirement: exception routing had to be built into the core product, not added on top. A 99% extraction rate is useless if the 1% exceptions end up in an unmanaged queue. Fieldiq routes exceptions to the right person in your organization — not back to a BPO.
The company is bootstrapped. We don't have a VC runway requiring us to expand into adjacent markets before we've solved this one. Document extraction, exception routing, and ERP integration — that's what we build.
The people building Fieldiq
Four people. ML document understanding, ERP integration, ops automation, and customer implementation. No one here came from a SaaS growth team.
Three things we refuse to trade away
We're hiring engineers who have worked on real ops problems.
Houston-based (hybrid), bootstrapped, four people. If you've worked in ML document understanding, ERP integration development, or enterprise back-office automation — and you want to work on a single focused product without a 200-person org — we'd like to talk.