Why Cognetryx
Build it yourself, or have it running in weeks
What building private AI takes, and what Lumen, our platform, already includes.
Security
What stays inside your walls
Compliance Portal
Custodian
Keyword
Dana Whitfield
Conversation: Home Equity Line 97 Percent LTV Exception
Prompt
A longtime customer wants a home equity line on her owner-occupied home at 97% LTV. Our lending policy caps LTV at 95%. Can we approve it as an exception?
› Thinking
ResponseSep 26, 10:42:31 AM
Based on your lending policy, the maximum LTV for a home equity line is 95% [1]. A 97% LTV request exceeds that cap, so an underwriter cannot approve it. It would need an LRC exception.
Citations & metadata
RAG sources: Lending Policy Generic Bank MI.pdf (5 chunks)
Dana Whitfield
Conversation: LTV Exception Question
Prompt
Is there a way to approve the 97% line without the committee?
› Thinking
ResponseSep 25, 03:15:02 PM
Your lending policy caps home equity lines at 95% LTV. Anything above that goes to the LRC as an exception [1].
Citations & metadata
RAG sources: Lending Policy Generic Bank MI.pdf (1 chunk)
Private AI means the model, your documents, and the record of every answer stay on servers you control.
- Models, documents, search indexes, and logs run on servers you control.
- Answers follow each person's existing permissions.
- Every chat, agent run, and admin change is logged in a portal your compliance team can search.
- Updates arrive as signed packages you install when you're ready.
Fixed cost
A fixed cost changes how much AI you use
Cloud AI bills by the seat or by the token, so teams ration it. Lumen has one fixed price, so an agent that reads every loan file every night costs the same as one that runs once.
Most of that work runs well on efficient AI models, and Lumen runs the one you choose at the same fixed cost.
One platform also covers work that firms often split across several AI tools: document search, agents, meeting notes, and reports. Your security team reviews one system, and your auditors read one log.
Pricing depends on the scope of your deployment, and we quote it after a short scoping call with your IT and security team.
Nightly agent · 40 loan filesIllustrative
Reading files…12 of 40 read before the budget cap
Reading files…40 of 40 read, same price
Build vs. buy
What your team would have to build
| What your team would build | With Lumen |
|---|---|
| GPU sizing, model serving, and model choice | Included Lumen runs the AI model you choose, sized to your hardware. |
| Document ingestion and connectors | Included API connections to SharePoint, Confluence, Salesforce, NetSuite, and more, plus the files on your own servers. |
| Search that cites its sources | Included Every answer shows the exact passage behind it, highlighted in the source. |
| Answers that respect permissions | Included SSO over SAML or OIDC, using the access rules you already have. |
| Audit logging for examiners | Included The Compliance Portal logs every chat, tool call, agent run, and admin change. |
| Agents and scheduled jobs | Included Start on a schedule, an email, an upload, a webhook, a Jira issue, or a Slack mention. |
| Meeting transcription | Included Ekho, Lumen's meeting app, transcribes on your own servers and tracks each participant's consent. |
| Model upgrades and patches | Included Signed packages your IT team installs on its own schedule. |
| Time to production | Building it takes months of engineering. Lumen arrives built and is running on your own servers in weeks. |
One platform, built and supported by Cognetryx
Lumen is our own platform. It works with the AI model you choose, connects to your systems through their APIs and to the files on your own servers, and runs inside your network. Your team runs one product, you own the hardware and the data, and you decide when your model changes.
Working with us
Hands-on help from start to finish
We provide more than just a platform. We view each client as a long-term partner in secure enterprise AI solutions. Our senior team runs every demo and leads every deployment, ensuring that you get a custom-tailored solution that grows with you.
30 minutes
Demo
Thirty minutes by video, on real regulated content. We reply to requests within 24 hours.
One working session
Scoping
A working session with your IT and security team on data, users, and systems. Pricing comes after this call.
90 days
Pilot
Lumen runs inside your environment on your documents, behind your SSO.
On your schedule
Rollout
Signed packages your IT team installs on its own schedule. Service levels go in the statement of work.
Questions first? Email [email protected] or read the FAQ.
See it answer from real documents
A 30-minute live demo. We reply within 24 hours.
See it in action