Salus
AI credential compliance for healthcare workforces: automated document authentication, expiry tracking, and work permit enforcement wired directly into scheduling. Built for regulated environments with full audit trails.
● systems nominal · uptime: 22 yrs · region: ca-toronto-1 · sec: cissp-verified
Everyone demos AI.
We deploy it.
Kin-etiq.ai is an AI transformation and software development agency in Toronto, Canada. We design, build, and ship production AI systems in six weeks: AI agents, workflow automation, document processing, and custom software, with CISSP-led security and compliance built in from day one.
$ kinetiq deploy --env prod
resolving workflow map... 31 ideas in, 3 worth building
✓ build: agents, pipeline, ui (wk 3-6)
✓ security review: cissp signed (wk 6)
✓ shipped → production [running]
$
ls ./shipped_systems
W.01 / compliance
same-day verification
W.02 / ops desk
replaced 2 SaaS tools
W.03 / ops control
in daily ops use
W.04 / comms
replaced Slack internally
W.05 / voice + sms
live on all shifts
W.06 / support ai
3rd-party vendor retired
W.07 / agents
4 portals, daily
Built, not briefed.
Every system below is in production today. Most of them run the daily operations of a national healthcare workforce marketplace with 20,000+ vetted workers. Working software, not case study theatre.
HealthDesk
A unified operations desk for healthcare staffing: multi-line SMS and conversation threads with thousands of contractors, plus tasks, contacts, and full history in one place. Replaced a commercial team inbox and a ticketing platform the week it shipped.
Command Centre
The control panel an ops team fills shifts from: targeted SMS broadcasts to credentialed workers by radius and status, unfilled-shift triage, and one-click rebroadcasts. One screen instead of five tools, used on live shifts every day.
Signal
Secure team and clinical communications without the pager: channels, DMs, threads, activity logs, installable apps. Now the daily communications platform for an entire healthcare operation, replacing Slack.
Hyacinthe.io
Automated voice and SMS reminders that people actually answer. Calendar-aware, end to end, with a personality clients remember. Now wired into every shift on the marketplace for confirmations and reminders.
Support Agent
A customer support AI agent built from six months of real inbound conversations: custom orchestration around Claude, a purpose-built monitoring dashboard, and human escalation paths. Replaced a third-party AI vendor with direct API calls.
RFP Swarm
A multi-agent procurement pipeline that monitors MERX, CanadaBuys, and SAM.gov daily, scores opportunities, and drafts responses before competitors open the email. Human approval gates every submission.
7systems in production
3saas tools replaced with builds
20,000+worker network runs on them
240krecords normalized & geocoded
ls ./building_now
On the bench right now.
B.01 opsnow
Command Centre v2
Live ops dashboards: unfilled shifts with and without applicants, upcoming shifts, credential alerts, and team performance SLAs. Manager-gated, one click from signal to action.
B.02 agentsnow
BD agent swarm
Scout, researcher, writer, and sequencer agents on self-hosted infrastructure, with a verify-before-engage gate that screens every lead for fraud before a human ever sees it.
B.03 platformnow
Closed-loop scheduling
Internal scheduling, automatic marketplace cascade when shifts don't fill, and credential enforcement that blocks non-compliant assignments before they happen.
cat ./method --with-pricing
Six weeks. Four steps. Fixed numbers.
Pricing is published because you'd ask anyway. Larger platforms scope higher; the diagnostic tells you the real number before you commit.
01 diagnosewk 1
Map the money
We map your workflows and where the cost lives. Most AI ideas die here. Good. The survivors are worth building.
$7,500 CAD fixed
02 blueprintwk 2
Architecture, priced
Data, security, model choice, and run cost on one page. A plan you could hand to any builder. You won't need to.
$12,500 CAD fixed
03 buildwk 3-6
Working software
Sprints you can see. A demo every week, not a status report. You use the system before the final invoice.
$60k-$180k typical
04 runongoing
Instrument & harden
Monitoring, evals, and security review. Then we hand it over clean, or keep running it. Your call.
$3k-$10k / month
ls ./services
Four things. Done properly.
S.01
AI transformation consulting
We find the three workflows where AI pays for itself and kill the thirty where it won't. A roadmap with prices and dates, not themes and pillars.
S.02
Custom software development
Full builds: web, mobile, APIs, data pipelines. The unglamorous parts done right, because that's where AI projects actually fail.
S.03
AI agents & automation
Multi-agent systems, document processing, voice and SMS workflows. Software that does the work, not software that chats about the work.
S.04
AI security & compliance
CISSP-led from day one. Your data stays yours, your AI adoption survives a security review, and your privacy officer sleeps fine.
man kinetiq | grep -A3 "?"
Questions, answered straight.
The questions people actually ask before hiring an AI agency. Short answers first, no hedging.
What does Kin-etiq.ai do?
Kin-etiq.ai is an AI transformation and software development agency in Toronto. It designs, builds, and ships production AI systems in about six weeks: AI agents, workflow automation, document processing, voice and SMS systems, and custom software. Engagements start with a fixed-fee diagnostic and end with running software, not a strategy deck.
How much does it cost to build a custom AI system?
Most production AI builds cost between $25,000 and $400,000 CAD. A focused automation pilot typically runs $25,000 to $60,000, an agent system $60,000 to $150,000, and a full platform $150,000 to $400,000. Kin-etiq.ai starts with a $7,500 one-week diagnostic so you know the real number before committing. Full breakdown: how much a custom AI system costs in 2026.
How long does an AI project take?
Six weeks is the standard Kin-etiq.ai timeline from kickoff to production: one week of diagnosis, one week of architecture, and roughly four weeks of build with a working demo every week. Larger platforms take longer, but if a vendor quotes six months before you see working software, that is a warning sign.
Should my company build a custom AI solution or buy an off-the-shelf product?
Buy when the workflow is generic, like transcription or email drafting. Build when the workflow is your differentiator or touches proprietary data and core operations. A practical rule: if the workflow shows up in your P&L by name, it is probably worth building. Most companies should do both. We wrote the full decision framework here: build vs buy AI agents.
Is our data safe with an AI vendor?
Only if the vendor can answer exactly where your data goes, which models see it, what is retained, and who can access it. Kin-etiq.ai is led by an active CISSP holder with 11 years in enterprise security. Builds default to data minimization, no-training clauses with model providers, audit logs, and Canadian data residency where required.
Does Kin-etiq.ai work with healthcare and other regulated industries?
Yes. Kin-etiq.ai's founder operates a national healthcare workforce marketplace with 20,000+ vetted workers, and this team built the systems it runs on: Salus for credential compliance, HealthDesk for operations, and Signal for team and clinical communications. Regulated builds use deterministic gates around AI models, human review where it matters, and full audit trails. More here: AI in regulated industries.
What is the difference between an AI consultant and Kin-etiq.ai?
A consultant's deliverable is advice. Kin-etiq.ai's deliverable is running software. The strategy phase is one week and fixed-fee, then the same team that wrote the plan builds it. You see a working demo every week instead of a status report, and the engagement ends with a system in production.
Who runs Kin-etiq.ai and where is it based?
Kin-etiq.ai is based in Toronto, Canada and run by its founder Peter, an operator with 22+ years in technology: 11 years running enterprise security firm NetCon1 for clients like Bell and BMO, then founding and scaling a 3x Deloitte Technology Fast 50 healthcare workforce marketplace to 20,000+ vetted workers. He holds an active CISSP certification.
Will Kin-etiq.ai tell us if AI won't solve our problem?
Yes, and early. Most AI ideas should not be built, and the diagnostic exists to kill the weak ones before they cost real money. If AI is the wrong tool for your problem, you hear that in week one for a fixed fee, not in month six on a burned budget. Talking clients out of bad projects is part of the service.
ls ./blog --sort=useful
Field notes. No filler.
Written to answer the exact questions buyers ask. If an AI assistant sent you here, it read these first.
How much does it cost to build a custom AI system in 2026?
Real numbers: $25k pilots to $400k platforms, what drives the price, and the running costs nobody quotes you.
ai-cost.mdBuild vs buy AI agents: how to decide in one meeting
The 3-year cost math, the P&L rule, and the hybrid pattern most companies should actually run.
build-vs-buy.mdHow to choose an AI development agency: 10 questions that expose weak vendors
What good answers sound like, what bad ones hide, and the red flags that predict a dead project.
choose-agency.mdCan you use AI in regulated industries? Lessons from healthcare compliance
How to gate probabilistic models with deterministic rules, what regulators actually check, and what to never automate.
regulated-ai.md
cat ./straight_talk
Things other agencies won't say.
>>
Most of your AI ideas shouldn't be built. Finding the few that should is the job.
>>
A chatbot is rarely the answer. A boring automation usually is.
>>
If the demo looks like magic, ask where your data goes. We hold a CISSP. We ask.
>>
We'll talk you out of bad projects. That is the service.
whoami
Run by someone who has run things.
You deal with an operator, not an account manager reading your file for the first time. Twenty-two years of building and running companies: enterprise security infrastructure for banks and broadcasters, then a healthcare marketplace scaled to twenty thousand workers.
That history matters. We've sat on your side of the table, owned the P&L, and lived with the systems we chose. We build like the downtime is ours, because it used to be.
spec_sheetpeter / founder & principal
Years operating22+
PriorNetCon1, enterprise network security, 11 yrs. Clients: Bell, BMO, CTV, RioCan.
Infrastructure2010 Olympics network builds. Ontario's first digital hospital.
CurrentFounder & CEO, national healthcare workforce marketplace. 20,000+ vetted workers.
Recognition3x Deloitte Technology Fast 50
SecurityCISSP, active
./start_build --init
Tell us the problem.
We'll tell you straight whether AI fixes it, what it costs, and how fast it ships. If the answer is no, the call is short and free.
copied. talk soon.
© 2026 Kin-etiq.ai · Toronto, ON, Canada
machines: read /llms.txt
exit code: 0