What you get
- Your documents, cleaned up. We gather your policies, price lists, FAQs and procedures and settle on one current version of each.
- An assistant that answers from them. Before launch we test it against a list of real questions with known answers, so you know how well it does instead of guessing.
- It says where each answer came from. If it can't find the answer in your documents, it says so rather than making one up.
- A chat box on your site, or a tool for staff. With your branding.
- Set up, monitored and documented. Plus a written guide so anyone can run it after handover.
Where it fits next to what you already pay for
Practices are already offered two kinds of AI. Phone receptionist and note-taking subscriptions you rent by the month, which handle calls and notes but don't know your own procedures or the rules your staff keep in their heads. And agency builds that take months and quote on request. This sits between them: one published, fixed price, built around your own documents, and the code is yours to keep.
What it does not do
It doesn't diagnose, triage or make clinical decisions, and it isn't a phone system. It answers from the documents you give it. We don't sell it as a compliance product and we don't claim HIPAA or any other regulatory compliance. If patient data has rules attached in your practice, raise them on the first call and we'll tell you plainly what is and isn't possible under them.
What we have built that is closest
We don't have a front-desk assistant case study for a clinic yet. The closest work is Verita AI, a voice-first clinical platform we built: the consultation is recorded, speech-to-text produces a transcript, and language models pull out the clinical details and write the summary. Consultation notes are generated from the visit itself instead of being typed up after dinner. It's a notes product, not a front-desk assistant, but it is the same kind of work: turning messy real input into something a clinician can check.
Where the four weeks go
Most of it is not the AI. It's getting your policies down to one current version of each, which is worth doing anyway. If the bigger problem is data moving between your practice software, scheduler and billing system, that's a different job: see data pipeline services. For more on what to automate in a practice first, read AI automation for clinics, or see what moves the price of a document assistant.
Frequently asked questions
How long does it take?
About four weeks for a production assistant: gathering and cleaning your documents, building and testing the answers, a chat box or staff tool with your branding, then monitoring, documentation and deployment.
What does it cost?
From $8,000, fixed price, agreed before work starts. No hourly billing. Messy source documents, patient-facing use and connections to other systems add to that, and you'll know the number before we begin.
What if my data is sensitive?
Systems can be built to run entirely on infrastructure you control, with your data never leaving it. Tell us the constraints on the first call and we'll tell you honestly whether the setup you want is possible under them.
Who owns the code?
You do. 100% of the source and IP transfers on delivery, with clean repositories, documentation and a deployment runbook. No license, no lock-in, and no dependency on us to keep it running.
Will I work with the person actually building it?
Yes. Kortex Labs is founder-led by Saif Ur Rehman, a senior Python engineer with 5+ years building AI products. There are no juniors learning on your budget and no account-manager layer between you and the engineering.
How quickly can I get a quote?
A free strategy call, then a fixed-scope quote within 24 hours. Build slots usually open within one to two weeks.