Yantra · Toronto, Canada
Your patient stops repeating a history you already hold. Your front desk stops re-typing it. Your clinicians stop finishing notes at nine at night. AI does the work behind the glass, and a person still signs.
Before anything touches a chart
Canadian data residency
Records and processing stay in Canada.
Privacy impact assessment first
Completed before any patient record is used.
PHIPA s.17 agent agreement
Written into the service agreement.
Zero data retention
Nothing is stored after processing or used to train models.
An audit log on every access
Every access to a record is logged.
A clinician approves
Nothing enters a chart without review.
These are commitments in every engagement, not badges on a wall. We are a studio, not a certified platform, so we will not show you a logo we have not earned. What we can put in writing is how the work is built, where your records live, who is accountable for it, and what goes in the contract before a single patient record is touched.
Who this is for
Independent practices that sign their own vendor contracts. Not hospitals, whose buying runs six months and whose pilots mostly never leave pilot. Find the sentence below that sounds like your Tuesday.
Documentation after hours
Phone lines nobody can answer · Referral letters See the segmentNo-shows and short-notice gaps
Treatment plan follow-up · Insurance predeterminations See the segmentIntake paperwork at the front desk
Progress notes between patients · Insurer and WSIB reporting See the segmentSession notes and treatment plans
Waitlist triage · Intake and consent paperwork See the segmentInternal referrals that never get booked
One patient, several separate charts · Who may see what, across disciplines See the segmentVisit scheduling and travel time
Care notes written in the car · Family communication See the segmentRequisition intake and triage
Report turnaround · Prior study retrieval See the segmentClaim rejections and resubmissions
Insurance verification · Coding accuracy See the segmentRefill and renewal requests by phone
Prescription intake and transfers · Blister packs and adherence follow-up See the segmentProducts
Each one is a single workflow, one number to move, and a person approving anything that counts, configured to your clinic in four to six weeks.
An independent test of the AI scribe your clinic already uses, on your own consented and de-identified visits, showing how often its notes are wrong and where.
Fixed-scope pilot
Sick notes, insurance and disability forms, and referral letters, drafted from the chart for the physician to review and sign.
Fixed-scope pilot
Phone messages, portal requests, faxes and walk-ins in one prioritized queue, with cancelled appointments offered to waiting patients. Staff make every triage decision.
Fixed-scope pilot
Fixed-scope pilot. Configured to your clinic and your systems. Not an off-the-shelf app yet, and we will say so until it is.
In the workshop
Yantrax Labs is a product studio. The eight offers on this site are real work and they fund the studio, and every one of them is also a week spent inside a clinic watching the same four things go wrong: a patient repeating their history, a fax being retyped, a note written at nine at night, a claim rejected three weeks late.
Those four are not eight different problems in eight different practices. They are one problem, and it is worth building something for. That is what we are doing with what the engagements teach us.
Built to the privacy file, not around it
Canadian residency, zero retention and an audit log on every access are the starting constraints, not a compliance pass bolted on before launch.
A person confirms anything that counts
Nothing charts, bills, or reaches a patient without someone releasing it. That rules out designs that demo better, and it is not negotiable.
Your practice system stays the source of truth
It reads and writes through the permissions your staff already have. Nobody is asked to migrate off the EMR they just finished paying for.
No launch date, because we have not earned the right to promise one. What you get instead is the working notes: what we are trying, what failed, and the first look when there is something real to look at.
One list, unsubscribe in a click. Your data stays yours, never shared or sold.
The person you would work with
"I have spent twelve years watching people fight software that was supposed to help them. Yantrax Labs is what I do about it."
More about the studioThe Yantrax principle
Most clinic “AI” staples a chatbot onto the same busy portal, and the patient still repeats their history at the desk. The work only changes at the intersection of the patient journey, the staff workflow and the model. So the step gets redesigned until nobody has to do it twice: the first win lands without replacing your EMR, proves itself in weeks, then goes deeper. You keep your practice system. It stops making people repeat themselves.
AI capability × great UX = productivity.
AI capability without UX is just an impressive demo. The same capability, engineered into a calm interface, is a tool people use every day.
The Yantrax difference
Drag the slider on each demo. Same job, re-imagined: fewer screens between your staff and the patient in front of them.
Intake, in one sentence
A clipboard of questions the chart already answers becomes one instruction.
Copilot drafted · review and confirm
The recall list, just ask
Four filters and an export become a plain-language question.
Foot checks · % of eligible screened
Built from one sentence. Every name is checkable in the chart.
Answers, not folder hunts
Asking whoever has been here longest becomes one sourced answer.
Urgent cardiology goes by fax to the regional intake line the same day, with the ECG attached, per the current pathway.
Answered while they are still at the desk · sources you can open
A patient is at the desk. You still are not sure which version is current.
Worked examples
Worked concepts, not client results. Each one takes an industry problem apart end to end.
A multidisciplinary clinic's cheapest appointment is the one a practitioner refers to a colleague thirty feet away, and it is the one most likely to evaporate. This is the transformation that catches the referral where it is made, carries only what the patient agreed to let travel, and escalates the ones that go quiet.
Untracked to 5 days
A community pharmacy answers the same question forty times a day, one interruption at a time, in the middle of dispensing. This is the transformation that lets the request arrive as structured intake instead of a phone call, keeps a pharmacist signing every fill, and measures the change at the counter rather than in a dashboard.
21.7%
Ontario pharmacy professionals naming workload as their top driver of burnout
A billing operation measures itself on how fast it reworks rejected claims, which rewards being efficient at a problem instead of ending it. This is the transformation that groups rejections by cause, sends the fix upstream to the clinic, and keeps a human on every resubmission.
$83B
Annual staff time spent on routine administrative transactions
Proof
Named, attributable feedback from people we have worked with. Not stock quotes, and the only thing on this site that counts as proof.
I've had the privilege of working with Kishan for about a year and a half. In that time he redesigned a massively complex problem with the user's perspective in mind. His professionalism and eye for detail makes him a terrific asset to any team. Within the year and a half that I worked with him, he designed the UI/UX for a learning management system, two different form builders, and various other screens to make the product at Variable successful.
Rick Caudill
CTO, Variable
Kishan is a passionate and thorough product designer. His ambition to always dive deeper and seek improvement was a real value to our org!
Brandon Gutierrez
Founder, BotStacks
Services
Clinic software goes wrong at the front desk, long before the architecture. So the hours get found first, then the fix is proved on your own forms, then it is built and released with the privacy file complete.
What to fix gets settled before anything is bought or built. Your intake, your booking, your notes and your claims get watched as they actually run on a normal day, then ranked by hours returned and revenue recovered. You leave with a decision you can defend to your partners, not a longer wish list.
You walk away with A one-page list of what to fix first
02The idea stops being a demo. We build the thing itself with Claude and the same tooling we build production software with, running against your own intake forms and your own note templates, so the front desk can try it on a Tuesday instead of picturing it.
You walk away with A working tool your staff can actually use
03The pilot becomes real: designed, engineered, privacy reviewed, tested and released. You watch it grow in an environment you can open any day, then it goes live, monitored, with a rollback that has been rehearsed on a working day rather than promised in a document.
You walk away with A live tool and the keys to it
04Rolling a working tool across locations takes more than one pair of hands, so it gets more: integration, data, mobile, whichever specialist the next site needs, brought in and led by the founder. You never get handed to an account manager.
You walk away with Every location on one standard
The process
Mature delivery, revolutionary outcomes. Four steps, fully collaborative, humans in the loop throughout.
We map your practice, your staff, and the busywork that drains them, then pinpoint where the journey costs your patient the most.
UX-led prototypes of the AI-native experience, with trust, control and transparency designed in from day one.
We ship production-grade Human + AI workflows, agents, copilots and generative interfaces, integrated with your systems.
Evals, metrics and iteration loops keep the product improving, with an eye on what is coming next.
The promises behind every step
The first win is the one that pays back fastest, proven in weeks. Then the ladder gets climbed rung by rung. No risky big-bang rewrite.
AI does the busywork while your people keep oversight. Every step is human-in-the-loop, with guardrails and reversible by design.
Your data never leaves your environment without your permission, and every build keeps it that way.
Insights
Who owns it once it is live, what you tell the board, and which problems AI is the wrong tool for. Written for the person who signs.
The clinic down the road can buy the same models you can, this quarter, at the same price. What they cannot buy is how it feels to be your patient. That is where the differentiation moved, and most clinic AI budgets are still aimed at the layer that is becoming commodity.
Read article
Ordinary software fails loudly and someone files a ticket. AI fails quietly and keeps answering. That difference is why a clinic AI system needs a named owner, a number, a cadence, and the authority to switch it off, and why most practices discover they have none of the four.
Read article
The same AI project reads as a threat or a plan depending entirely on how it is announced. Four facts make the difference, and they are the same four at the partners' meeting, at the staff huddle, and in anything a patient ever sees.
Read articleYantrax Labs is the Toronto-based healthcare experience design studio of founder Kishan Thankey. It works with independent and allied practices: family medicine, dental, physiotherapy, mental health, home care, imaging and billing. The journey your patient takes and the work your staff do around it get mapped and redesigned end to end: fewer steps, shorter waits, and nothing anyone has to type twice. AI is how those steps get removed, not the thing being sold.
Both, because the experience breaks in both places. A journey is mapped, the steps that cost your patient and your front desk time are redesigned, and the software behind them is built or rebuilt so the new experience holds up in a real clinic on a real Monday.
AI is the method, not the product. Once the journey shows where your patient waits or repeats themselves, and where your staff re-type what someone already said, AI is usually what removes those steps: reading the requisition, drafting the note, pulling the prior result. The patient never has to know it is there.
Yantrax Labs is based in Toronto, Canada, and works with healthcare practices across Canada. Ontario work is designed against PHIPA from the start, including Canadian data residency and a privacy impact assessment completed before deployment.
A process where the reading, drafting and looking-up happen automatically, and a person still makes every decision that carries consequence. Nothing charts, bills or reaches a patient without a clinician or staff member confirming it, and every step is explainable, reversible and overridable.
Yes, when the old system is what your patient is actually waiting on. Aging codebases (old PHP, C/C++, legacy .NET or Java) get migrated with tests that lock in current behavior, so the experience can change without a risky big-bang rewrite. Your EMR or practice management system is not replaced; it stays the source of truth.
Your data stays yours. Personal health information stays in Canadian infrastructure and never leaves your environment without your permission. Every solution runs against your own systems with grounding, guardrails, audit logs on every access, and zero data retention in processing, so intelligence gets added without giving your patients away.
The 14 places a practice loses hours inside software it already pays for, and what each becomes when a step is removed. Every number in it is sourced.
You get the occasional post by email too. Your data stays yours, never shared or sold.
A free, no-obligation consult. One clinic workflow gets mapped and ranked by what it pays back, and patient information never leaves Canada.
The founder is on every engagement. Specialists join for the stage that needs them, never an account manager.