Offline AI for professional practices

AI assistance for your practice. Client and patient data never leaves the building.

An offline workstation that drafts from the facts your team enters, answers questions from your own documents, and looks up your procedures — running entirely on a computer in your office. No internet required, nothing sent to the cloud.

Request a demo Choose your business

Nothing is sent to the cloud Works with the internet unplugged Your team reviews every draft

The EdgePlus AI workstation: a compact computer on an office desk, running the practice's AI locally — The workstation sits in your office and runs everything locally — the model, your documents, and the search index.

Choose your business

Same offline platform, tailored to your profession — your document pack, your templates, and an evaluation graded on your own material. Pick your business to see how it works for you.

Dental practices

Dentists

Claim narratives, appeal letters, CDT coding answers, and protocol lookup for dental offices — the vertical where EdgePlus AI was built and proven.

Proof of concept complete

Veterinary practices

Veterinarians

SOAP and exam note drafts, discharge instructions, and pet-insurance narratives — with medical records and client data staying in the building.

Founding pilots now recruiting

Accounting & tax firms

Accounting

Client letters, notice responses, and sourced research answers — with client financial data and PII never leaving the office.

Founding pilots now recruiting

Small law firms

Legal

Client correspondence, routine drafts, and cited memos from your own sources — with client confidences never sent to a cloud AI.

Founding pilots now recruiting

What it does

The paperwork side of a professional practice — drafting, questions, and lookups — handled by a system that works only from your documents and the facts your team enters.

Drafts from your facts

Letters, narratives, notes, and routine documents drafted from a required facts form your team fills in first. The system drafts only from those facts — and your team reviews every draft before it goes anywhere.

Answers from your documents

Professional questions answered from reference material loaded on the workstation — your codes, publications, and precedents — with the source shown so staff can verify in seconds.

Your procedures, on demand

Your own SOPs, policies, and templates answered in plain language, with the document the answer came from named. New staff get the office's way of doing things, not a generic guess.

Private by architecture

The model, your documents, and the index all live on one computer in your office. There is no cloud account and nothing is transmitted out — privacy is a property of the setup, not a policy promise.

How it works

Three parts, all on one computer in your office. No cloud account, no subscription service, nothing transmitted out.

Part one

The writer

A local AI model installed on the workstation. It reads the material it's given and drafts in plain, professional language. It runs on the machine itself, not on a server somewhere else.

Part two

The binder

Your practice's own documents and references: SOPs and protocols, office policies, templates, and professional references for your field. This is the only knowledge the system answers from — your documents supply the requirements, your team supplies the facts.

Part three

The index

Search that finds the right pages in the binder and hands them to the writer before it answers. Answers cite the document they came from, so staff can check any answer against the source in seconds.

Everything runs on the box in your office.

Retrieval, drafting, and citation were all tested and verified with the internet disconnected. If your connection goes down, the workstation doesn't notice.

Proven in our first vertical — dentistry

EdgePlus AI has completed a full proof of concept in dentistry, graded test by test. That proof validates the platform and the method. Each new vertical gets its own document pack and its own graded evaluation — we don't borrow results across professions.

What the dental proof of concept showed

Narrative, appeal, and coding tests passed when the chart facts were supplied — including a root canal narrative with no invented clinical findings, whose two framing errors were caught in staff review.
Document retrieval verified — questions about office protocols answered from the practice's own documents, with the source named.
Reference gaps fixed by documents — adding a reference chart to the binder corrected answers the model had previously gotten wrong from memory.
Every function tested offline — retrieval, drafting, and citation all worked with the internet unplugged.

The test it failed — and the design answer

In dentistry we also tested the system the hard way: asked to draft given a case reference and no facts at all, the model invented a complete, plausible, fictional draft. Stronger instructions did not stop it.

So the product doesn't rely on instructions. Drafting is gated behind a required facts form in every vertical: your team enters the facts first, the system drafts only from those facts, and your team reviews every draft before use.

Where each vertical stands: Dentistry — proof of concept complete, trials under way. Veterinary, accounting, and legal — the same platform, now recruiting founding pilots. Each pilot is scored against measures agreed before day one, on that practice's own documents.

The 30-day trial

A structured month in your office, evaluated against measures agreed with you before day one — not against our promises. The shape is the same in every vertical; the documents and test set are yours.

Day 1

Setup and baseline

A half-day install. We capture baseline numbers — how long the drafting work currently takes — and read every setting back on screen together, so the configuration is verified, not assumed.

Days 1–2

Your documents, loaded and tested

We load your office's own SOPs, policies, and templates, and build a graded test set from your documents, so accuracy is measured on your material.

Day 2

Team session

A 30-minute session on the workflow: facts form → draft → review. What the system does, what it never does, and where human review sits.

Days 3–30

Live use, with a simple log

The workstation is used in real work. Your team keeps a simple log of drafts used, corrections made, and time taken — the raw material for an honest evaluation.

Day 30

Evaluation meeting

We score the trial together against the measures agreed before day one: accuracy on a test set built from your own documents, draft fidelity — zero invented findings in accepted drafts — time saved against the Day-1 baseline, and your team's verdict.

Your Business / Dentists

Offline AI for dental practices

AI assistance for your practice. Patient data never leaves the building.

An offline workstation that drafts claim narratives and appeals, answers CDT coding and documentation questions, and looks up your office's own protocols — running entirely on a computer in your office. No internet required.

Request a demo See how it works

Nothing is sent to the cloud Works with the internet unplugged Staff review every draft

What it does

The paperwork side of the practice — drafting, coding questions, and protocol lookups — handled by a system that works only from your documents and the facts your staff enter.

Claim narratives

Drafted from the chart facts your staff enter into a required facts form — tooth, findings, existing restoration, plan. The system drafts only from those facts, and staff review every draft before it goes anywhere.

Appeal letters

Built from the denial and the chart facts, answering the payer's stated reason point by point, with the supporting attachments listed — such as the dated radiograph enclosed with the appeal.

Coding & documentation answers

CDT questions — the difference between D4341 and D4342, what documentation each code requires — answered from reference material loaded on the workstation, with the source shown.

Office protocol lookup

Your practice's own SOPs and policies — sterilization protocol, front-desk and billing policies, aftercare sheets — answered in plain language, with the document the answer came from named.

How it works

Three parts, all on one computer in your office. No cloud account, no subscription service, nothing transmitted out.

Part one

The writer

A local AI model installed on the workstation. It reads the material it's given and drafts in plain, professional language — claim narratives, appeal letters, and answers to questions. It runs on the machine itself, not on a server somewhere else.

Part two

The binder

Your practice's own documents and references: SOPs and protocols, office policies, aftercare sheets, coding references. This is the only knowledge the system answers from — your documents supply the requirements, your staff supply the facts.

Part three

The index

Search that finds the right pages in the binder and hands them to the writer before it answers. Answers cite the document they came from, so staff can check any answer against the source in seconds.

Everything runs on the box in your office.

Retrieval, drafting, and citation were all tested and verified with the internet disconnected. If your connection goes down, the workstation doesn't notice.

Proven, and honest about its limits

EdgePlus AI has completed a full proof of concept on a laptop workstation, graded test by test. Here is what it did — including the test it failed, and what that failure changed in the design.

What the proof of concept showed

Narrative, appeal, and coding tests passed when the chart facts were supplied — including a root canal narrative with no invented clinical findings, whose two framing errors were caught in staff review.
Document retrieval verified — questions about office protocols answered from the practice's own documents, with the source named.
Reference gaps fixed by documents — adding a tooth-numbering chart to the binder corrected answers the model had previously gotten wrong from memory.
Every function tested offline — retrieval, drafting, and citation all worked with the internet unplugged.

The test it failed — and the design answer

We also tested the system the hard way: asked to draft a crown narrative given a tooth number and no facts at all, the model invented a complete, plausible, fictional narrative — findings, decay, radiographic evidence, even patient agreement — none of it real. Stronger instructions did not stop it.

So the product doesn't rely on instructions. Drafting is gated behind a required facts form: staff enter the chart facts first, the system drafts only from those facts, and staff review every draft before use. The model never sees an empty page — and a fact it wasn't given can't appear in the draft.

Scope, stated plainly: EdgePlus AI handles documentation, coding questions, administration, and patient education materials. It does not diagnose, and it does not recommend treatment — those decisions stay with the dentist, where they belong.

The 30-day trial

A structured month in your office, evaluated against measures agreed with you before day one — not against our promises.

Day 1

Setup and baseline

A half-day install. We capture baseline numbers — how long narratives and appeals currently take — and read every setting back on screen together, so the configuration is verified, not assumed.

Days 1–2

Your documents, loaded and tested

We load your office's own SOPs and policies — sterilization protocol, front-desk and billing policies, aftercare sheets — and build a graded test set from your documents, so accuracy is measured on your material.

Day 2

Staff session

A 30-minute session with your team on the workflow: facts form → draft → review. What the system does, what it never does, and where human review sits.

Days 3–30

Live use, with a simple log

The workstation is used in real work. Staff keep a simple log of drafts used, corrections made, and time taken — the raw material for an honest evaluation.

Day 30

Evaluation meeting

We score the trial together against the measures agreed before day one: accuracy on a test set built from your own documents, draft fidelity — zero invented findings in accepted drafts — time saved per narrative and appeal against the Day-1 baseline, claim outcomes, and your staff's verdict.

Your Business / Veterinarians

Offline AI for veterinary practices

Clinical paperwork, drafted in your practice. Records and client data stay in the building.

An offline workstation for independent veterinary practices that drafts exam notes, discharge instructions, and insurance narratives from the facts your care team enters — running entirely on a computer in your practice. No internet required.

Ask about the founding pilot See how it works

Medical records never sent to the cloud Works with the internet unplugged Care team reviews every draft

A veterinary examination room with an exam table and the EdgePlus workstation on the counter — The workstation sits in your practice and runs everything locally — the model, your documents, and the search index.

What it does

Clinical documentation and client paperwork — drafted from the facts your care team enters and the templates and protocols your practice already uses.

SOAP & exam note drafts

Structured note drafts built from the facts the care team enters into a required facts form — signalment, history, findings, assessment, and plan as supplied. The system drafts only from those facts, and the care team reviews every note.

Discharge & aftercare instructions

Client-ready discharge and home-care instructions generated from your practice's own templates and the procedure facts entered for that patient — consistent wording, every time, reviewed before it goes home with the client.

Pet-insurance claim narratives

Narratives drafted from chart facts — dates, findings, and treatment as recorded — answering the insurer's questions from the record, with staff review before submission.

Protocol & policy lookup

Your practice's own SOPs and policies — anesthesia monitoring, isolation, front-desk and billing policies — answered in plain language, with the document the answer came from named.

How it works

Three parts, all on one computer in your practice. No cloud account, no subscription service, nothing transmitted out.

Part one

The writer

A local AI model installed on the workstation. It reads the material it's given and drafts in plain, professional language — notes, instructions, narratives, and answers. It runs on the machine itself, not on a server somewhere else.

Part two

The binder

Your practice's own documents: SOPs and protocols, discharge templates, aftercare sheets, and office policies. This is the only knowledge the system answers from — your documents supply the requirements, your care team supplies the facts.

Part three

The index

Search that finds the right pages in the binder and hands them to the writer before it answers. Answers cite the document they came from, so staff can check any answer against the source in seconds.

Everything runs on the box in your practice.

The platform's retrieval, drafting, and citation were tested and verified with the internet disconnected. If your connection goes down, the workstation doesn't notice.

Where veterinary stands — stated plainly

No proof of concept has been run for veterinary yet, and we won't pretend otherwise. Our proof of concept was completed in dentistry; veterinary is in pilot recruitment. What transfers is the platform and the method — what has to be earned, in your practice and on your documents, is the veterinary result.

The method, proven in dentistry

Offline operation verified — retrieval, drafting, and citation all worked with the internet unplugged.
Documents cure knowledge gaps — adding a reference to the binder corrected answers the model had gotten wrong from memory.
Drafting gated by design — the facts-form gate exists because dental testing showed instructions alone don't stop a model from inventing facts it wasn't given.

What a founding pilot does

Founding veterinary pilots run the same graded method: we load your practice's own SOPs, templates, and policies, build a test set from your documents, and score the system against measures agreed with you before day one.

Until a veterinary pilot completes, there are no veterinary results to quote — no accuracy figures, no time savings, no testimonials. If a vendor shows you those numbers for a product that hasn't run in a practice like yours, ask where they came from.

Scope, stated plainly: EdgePlus AI handles documentation, administration, and client education materials. It does not diagnose, and it does not recommend treatment — those decisions stay with the veterinarian, where they belong.

The founding-pilot 30-day evaluation

A structured month in your practice, evaluated against measures agreed with you before day one. Founding pilots help prove the vertical — and get the honest, graded answer about whether it works on your documents.

Day 1

Setup and baseline

A half-day install. We capture baseline numbers — how long notes, discharge instructions, and narratives currently take — and read every setting back on screen together.

Days 1–2

Your documents, loaded and tested

We load your practice's own SOPs, templates, and policies and build a graded test set from your documents, so accuracy is measured on your material.

Day 2

Team session

A 30-minute session on the workflow: facts form → draft → review. What the system does, what it never does, and where human review sits.

Days 3–30

Live use, with a simple log

The workstation is used in real work. Your team keeps a simple log of drafts used, corrections made, and time taken.

Day 30

Evaluation meeting

We score the pilot together against the measures agreed before day one: accuracy on your test set, draft fidelity — zero invented findings in accepted drafts — time saved against baseline, and your team's verdict.

Your Business / Accounting

Offline AI for accounting & tax firms

Client correspondence and research, drafted in your office. Client financial data never leaves it.

An offline workstation for small accounting and tax firms that drafts client letters and notice responses from supplied facts, and answers research questions from loaded references with the source named — running entirely on a computer in your office.

Ask about the founding pilot See how it works

Client PII never sent to the cloud Fits IRS & FTC Safeguards expectations Figures only from supplied facts

A small accounting office desk with the EdgePlus workstation, reference binders, and filing cabinets — The workstation sits in your office and runs everything locally — the model, your references, and the search index.

What it does

Client correspondence, notice responses, and sourced research — drafted from the facts and documents your firm supplies, and reviewed by your professionals before anything is sent.

Client letters & emails

Routine client correspondence — engagement follow-ups, document requests, deadline reminders, explanations in plain language — drafted from the facts your team enters into a required facts form, in your firm's tone and templates.

Notice response drafts

IRS and state notice responses built from the notice itself plus the client's facts as supplied — structured point by point against what the notice asks, ready for professional review before filing.

Research answers, with sources

Questions answered from loaded references — IRS publications, firm procedures, and your own research library — with the source named for every answer, so the professional can verify before relying on it.

Procedure & policy lookup

Your firm's own procedures — onboarding, document handling, review checklists, deadlines — answered in plain language from your internal documents, with the source named.

How it works

Three parts, all on one computer in your office. No cloud account, no subscription service, nothing transmitted out.

Part one

The writer

A local AI model installed on the workstation. It reads the material it's given and drafts in plain, professional language — letters, responses, and research answers. It runs on the machine itself, not on a server somewhere else.

Part two

The binder

Your firm's references and procedures: IRS publications, state guidance you rely on, firm procedures, and letter templates. This is the only knowledge the system answers from — your documents supply the requirements, your team supplies the facts.

Part three

The index

Search that finds the right pages in the binder and hands them to the writer before it answers. Answers cite the document they came from, so your team can check any answer against the source in seconds.

Everything runs on the box in your office.

The platform's retrieval, drafting, and citation were tested and verified with the internet disconnected. Client data that never leaves the office is data that can't leak through a cloud service — a posture that fits IRS and FTC Safeguards expectations for firms handling taxpayer information.

Where accounting stands — stated plainly

No proof of concept has been run for accounting yet, and we won't pretend otherwise. Our proof of concept was completed in dentistry; accounting is in pilot recruitment. What transfers is the platform and the method — the accounting result has to be earned on your firm's documents.

The method, proven in dentistry

Offline operation verified — retrieval, drafting, and citation all worked with the internet unplugged.
Answers tied to loaded sources — reference material in the binder corrected answers the model had gotten wrong from memory.
Drafting gated by design — the facts-form gate exists because dental testing showed instructions alone don't stop invention.

Figures discipline — and the founding pilot

Numbers come only from supplied documents and facts — never generated. A draft may only state a figure your team entered or that appears in a loaded source; if a figure isn't supplied, the draft leaves it out rather than estimating one.

Founding accounting pilots run the same graded method as dentistry: your references and procedures loaded, a test set built from your documents, scored against measures agreed before day one. Until a pilot completes, there are no accounting results to quote.

Scope, stated plainly: EdgePlus AI provides drafting and research support. It does not prepare returns, give tax advice, or exercise professional judgment — those stay with the accountant, where they belong.

The founding-pilot 30-day evaluation

A structured month in your firm, evaluated against measures agreed with you before day one.

Day 1

Setup and baseline

A half-day install. We capture baseline numbers — how long client letters, notice responses, and research currently take — and read every setting back on screen together.

Days 1–2

Your documents, loaded and tested

We load your references, procedures, and templates and build a graded test set from your documents, so accuracy is measured on your material.

Day 2

Team session

A 30-minute session on the workflow: facts form → draft → review — including the figures discipline: numbers only from supplied facts.

Days 3–30

Live use, with a simple log

The workstation is used in real work. Your team keeps a simple log of drafts used, corrections made, and time taken.

Day 30

Evaluation meeting

We score the pilot together against the measures agreed before day one: accuracy on your test set, draft fidelity — zero invented figures or findings in accepted drafts — time saved against baseline, and your team's verdict.

Your Business / Legal

Offline AI for small law firms

Drafting help that keeps client confidences in the office.

An offline workstation for small law firms that drafts client correspondence and routine documents from facts your legal team supplies, and summarizes loaded source material with citations — running entirely on a computer in your office. No cloud AI ever sees a client confidence.

Ask about the founding pilot See how it works

No cloud AI sees client confidences Every citation traces to a loaded source Every draft is attorney-reviewed

A small law firm office with bookshelves and the EdgePlus workstation on the desk — The workstation sits in your office and runs everything locally — the model, your documents, and the search index.

What it does

Drafting load and research verification — handled from the facts your legal team supplies and the sources loaded on the workstation, with attorney review on everything.

Client correspondence

Status updates, document requests, and routine client letters drafted from the facts the legal team enters into a required facts form — in the firm's voice, reviewed by an attorney before sending.

Routine document drafts

First drafts of routine documents assembled from supplied facts and the firm's own templates and precedents — the attorney starts from a structured draft instead of a blank page, and finishes it.

Memos with citations

Memos summarizing loaded source material — research files, records, and references on the workstation — with every point cited to the source it came from, so verification is a check, not a re-do.

Procedure & precedent lookup

The firm's own procedures, checklists, and precedent library answered in plain language from your documents — how this firm does it, found in seconds, with the source named.

Intake and status letter drafting is part of the correspondence workflow above: intake summaries and status letters are drafted from the information your team enters, never from invented matter details.

How it works

Three parts, all on one computer in your office. No cloud account, no subscription service, nothing transmitted out.

Part one

The writer

A local AI model installed on the workstation. It reads the material it's given and drafts in plain, professional language — correspondence, routine documents, and memos. It runs on the machine itself, not on a server somewhere else.

Part two

The binder

Your firm's own documents: templates, precedents, procedures, and the source material you choose to load. This is the only knowledge the system answers from — your documents supply the requirements, your legal team supplies the facts.

Part three

The index

Search that finds the right pages in the binder and hands them to the writer before it answers. Answers cite the document they came from, so an attorney can verify any point against the source in seconds.

Everything runs on the box in your office.

The platform's retrieval, drafting, and citation were tested and verified with the internet disconnected. Client confidences are never typed into a cloud service, because there is no cloud service in the loop.

Where legal stands — stated plainly

No proof of concept has been run for legal yet, and we won't pretend otherwise. Our proof of concept was completed in dentistry; legal is in pilot recruitment. What transfers is the platform and the method — the legal result has to be earned on your firm's documents.

The method, proven in dentistry

Offline operation verified — retrieval, drafting, and citation all worked with the internet unplugged.
Citations tied to loaded sources — answers named the document they came from, and reference documents corrected answers given from memory.
Human review caught real errors — in the dental POC, staff review caught framing errors in an otherwise sound draft. Review is part of the product, not an afterthought.

Verification is the design, not a disclaimer

Every draft is attorney-reviewed and every citation traces to a loaded source. A citation the system gives can be opened and checked against the document it claims to come from.

The facts-form gate exists because the dental testing showed instructions alone don't stop invention: asked to draft with no facts supplied, the model invented a complete fictional narrative, and stronger wording didn't prevent it. So drafting is structurally gated — no facts entered, no draft — in every vertical, legal included. Founding pilots are scored against measures agreed before day one; until one completes, there are no legal results to quote.

Scope, stated plainly: EdgePlus AI provides drafting and research support for legal professionals. It is not legal advice to the public, and it does not replace attorney judgment — every output is reviewed by an attorney before use.

The founding-pilot 30-day evaluation

A structured month in your firm, evaluated against measures agreed with you before day one.

Day 1

Setup and baseline

A half-day install. We capture baseline numbers — how long correspondence, routine drafts, and memo work currently take — and read every setting back on screen together.

Days 1–2

Your documents, loaded and tested

We load your templates, precedents, and procedures and build a graded test set from your documents, so accuracy and citation fidelity are measured on your material.

Day 2

Team session

A 30-minute session on the workflow: facts form → draft → attorney review. What the system does, what it never does, and how citations are verified.

Days 3–30

Live use, with a simple log

The workstation is used in real work. Your team keeps a simple log of drafts used, corrections made, and time taken.

Day 30

Evaluation meeting

We score the pilot together against the measures agreed before day one: accuracy on your test set, citation fidelity, draft fidelity — zero invented facts in accepted drafts — time saved against baseline, and your team's verdict.