When your founder retires,
your edge goes with them.
Sentinel is the Apex Replicant product tuned for SMBs and funded startups facing key-person risk. The institutional expert inside your company (founder, master craftsman, senior partner) gets a Sentinel-branded protege trained on their methodology. New hires onboard against it. The organization keeps the knowledge when that person retires, leaves, or scales away.
The knowledge that walks out the door.
The founder with ten years of customer-discovery instinct doesn't write it down. The master craftsman who can read a job in 30 seconds doesn't document the read. The senior partner whose judgment closes the deal doesn't have a methodology doc to hand to the new hire.
That tacit knowledge is what makes the company. And it walks out the door the day they do.
Every knowledge-management tool in this category is text-based: Q&A, search, summaries. Sentinel captures what those tools can't: the live voice of the person who built the business, available to your team on demand.
How it works.
Structured interview
We sit down with your institutional expert across a series of guided voice sessions. The Apex Replicant platform records, transcribes, and structures the methodology that lives in their head.
Voice training
The protege is trained on their voice, their philosophy, their decision rules, and the way they actually talk about the work. We ship the first version in under 24 hours from the final interview.
Ongoing refinement
Every new conversation a team member has with the Sentinel sharpens it. The institutional expert can drop in feedback any time ("be less formal", "always reference the post-mortem framework") and the protege updates itself in one request.
Why Sentinel beats text-first knowledge tools.
- Glean searches the docs you have already written.Sentinel captures the methodology that was never in a doc.
- Notion AI answers from whatever has been typed in.Sentinel answers in the founder's actual voice, on a live call.
- Guru "verified answers" are human-confirmed text.Sentinel's verification is the expert refining the protege in real time.
- Custom GPTs need a week of dictation plus a prompt engineer.Sentinel ships in under 24 hours after the final interview.
Who Sentinel is for.
Funded Founders
Seed-to-Series-B startups ($1M to $10M raised, 5 to 30 employees) where founder methodology IS the company’s edge and new hires can’t onboard fast enough.
New Businesses
Recently-incorporated companies where founder bandwidth is the rate-limiter on hiring leverage.
Professional services firms
10 to 100 employees, where the senior partner’s judgment closes 80% of the deals.
Specialty trades
Master craftsmen whose skill the next generation doesn’t yet have.
Healthcare practices
A senior partner approaching retirement; the methodology is the practice.
Family-owned manufacturing
The founder still in the building, decades of judgment unwritten.
Live capture workshops (services upsell).
For companies with an expert approaching retirement or transition, we run live capture workshops: a structured one-to-two-day engagement to extract and voice-record the methodology that defines the practice. $10K to $25K depending on scope, included in onboarding for Enterprise customers.
Talk to us about a capture workshop: mailto:ai@expertscale.ai?subject=Sentinel%20capture%20workshop
Pricing.
Sentinel runs on the same voice infrastructure that powers every protege on Apex Replicant. The $40/hour rate is the rack rate for outbound voice usage; we pool credit hours into each plan, with overage billed at the rate appropriate to the tier.
- Pilot
Pilot
$1,500/mo1 Sentinel, up to 20 employee accounts, 25 hrs/mo voice
Best for: Companies validating Sentinel with one institutional expert and a small onboarding cohort.
- 1 Sentinel-branded protege
- Up to 20 employee accounts
- 25 hours per month voice usage
- Structured interview + voice training
- Refinement loop (auto-regen + AI-surfaced suggestions)
Effective: ~$60/hr blended (Pilot has compressed onboarding)Overage: $42/hrOnboarding: $3,500 (compressed for pilot) - Most Popular
Growth
$3,500/mo3 Sentinels, up to 100 employee accounts, 75 hrs/mo voice
Best for: Multi-team operations where several institutional experts each need their own Sentinel.
- Up to 3 Sentinels
- Up to 100 employee accounts
- 75 hours per month voice usage
- Audit trail exports
- 4D Knowledge Scoping (per-org, per-Sentinel, per-user, per-session)
- Standard onboarding workshop
Effective: ~$47/hr blendedOverage: $40/hr (rack)Onboarding: $5,000 (standard) - Enterprise
Enterprise
$8,000+/moUnlimited Sentinels + employees, 200 hrs/mo voice, SSO, SOC 2 Type II readiness, dedicated success
Best for: Multi-site organizations with compliance review and a portfolio of senior experts to preserve.
- Unlimited Sentinels and employee accounts
- 200 hours per month voice usage
- Single sign-on (SSO)
- SOC 2 Type II readiness controls + audit-trail exports
- Dedicated customer success
- Custom data residency on request
Effective: $40/hr blendedOverage: $36/hr (10% tier discount)Onboarding: $10,000 to $25,000 (white-glove)
Annual prepay: 17% off plus bonus credit pool (~15% extra hours). SOC 2 Type II is in audit; full attestation expected Q3 2026.
Common questions.
- What's the difference between Sentinel and Glean / Notion AI?
- Glean and Notion AI search what your team has already written down. Sentinel captures what your senior expert never wrote down: the live, voice-based judgment they use every day. Different problem, different mechanism.
- How long until our team can talk to a Sentinel?
- The first Sentinel ships in under 24 hours after the final structured interview. The structured-interview process itself runs across 1 to 5 sessions depending on the depth of methodology being captured.
- What happens if our institutional expert leaves before we finish?
- That's the failure mode Sentinel exists to prevent. The structured-interview process is designed to capture methodology in phases: even a single interview produces a usable v1 Sentinel. The longer the runway, the deeper the capture.
- Is our data isolated from other Sentinel customers?
- Yes. Sentinel uses our 4D Knowledge Scoping pattern: every retrieval is scoped by company, Sentinel, user, and session. Hard isolation enforced at retrieval time. Enterprise tier adds SOC 2 Type II readiness controls and optional custom data residency. See /security for the full architecture.
- Do new hires need to install anything?
- No. New hires talk to the Sentinel through a voice widget on your internal portal, or via a phone number we provision for your company. Browser plus microphone is the only requirement.
Talk to us about
your Sentinel.
The founder isn't getting younger. The methodology isn't getting written down. Sentinel is the bridge.