Charm builds and runs outbound for
+ others
+ others
Enginius was built the hard way: five years on a Department of Defense contract, a sovereign language model rather than a wrapper, deployed behind the customer's firewall where every output is traceable and deterministic. Since commercialization opened in early 2026, four medical device makers have signed, every one of them through your board and your references.
References are the best channel there is, and they do not run on a schedule. Getting from four customers to twenty takes a channel you control. Your team already built the sensors: agents that hear regulatory and leadership changes across the market. What is missing is your own phrase for it, the art of making it a conversation. This document is that conversation engine: who it talks to, what it says word for word, and what it costs.
+ others
Regulated industries will never trade control and context for capability. Your words, and they are the whole pitch. A sovereign model deployed behind the customer's firewall, on a single GPU, with no data leaving the building, is the only architecture a medical device quality organization can say yes to without a fight. Generic AI is disqualified before the demo starts. You are not.
A regulatory submission document that takes a team four months to draft, and your platform returns a first draft in under 30 seconds, from their own design history files and the standards themselves. Cold email lives and dies on one concrete sentence, and you own one of the best we have ever been handed.
Every output traceable, referenceable, and the same every time. A submission your customer can still defend in an audit ten years out, for devices that live for fifty. Where the rest of the AI market apologizes for hallucination, your architecture simply does not do it, and the buyer this proposal targets knows exactly why that matters.
Medical device regulatory life is published as it happens: clearances, submissions, recalls, enforcement actions, MDR deadlines, leadership changes. Every one of those is a dated, public reason to write to a named person this week. This is the best signal terrain we work in, and it is the terrain your own agents already listen to.
Funded by the U.S. Army, locked from commercializing until the work shipped, deployed behind Philips Healthcare's firewall on a seven-year-old GPU. Four paying medical device customers within months of opening the doors. The story survives the skeptical read, which in this vertical is the only read there is.
Board introductions and industry references produced four excellent customers, and they will keep producing occasionally, on nobody's calendar. A Series A window at the end of this year does not wait for occasionally. Four to twenty is a volume of conversations the current channel cannot schedule, and everyone on your side already knows it.
Your team built Claude-code agents that track regulatory and leadership changes across the market. That puts you ahead of most companies we meet. But a flagged signal is not a booked meeting, and by your own account the missing piece is the art of executing on it and making it a conversation. That art is the entire business we are in.
Half your team are rocket scientists and half are data scientists, and every hour they spend on prospecting lists and outreach copy is an hour off the platform that wins the deals. The engine has to run without borrowing the people who build the product.
Regulatory and quality leaders are the most hallucination-suspicious readers in B2B, and most AI vendors write to them in generated-sounding copy stuffed with claims. The outreach has to clear the same bar the product does: specific, referenced, and honest about what it is. Copy that smells automated does not just fail here, it damages the name on it.
You want one or two hires now and a path to twenty. Hire them into silence and you pay salaries while they guess at the pitch. The engine's first job is finding the message that converts, so that every seat you add afterward inherits a script that already works.
First, straight volume. Every US medical device establishment is on public record, and the regulatory, quality and engineering leaders inside them are reachable by name. The plain offer, said clearly: first drafts of submission and quality documents from your own files, behind your firewall, deterministic. Sometimes the winning campaign is simply that sentence delivered to everyone it is true for, at full volume from day one.
Second, the signals. Clearances published weekly, recalls and enforcement actions as they land, MDR deadlines, leadership changes in regulatory and quality seats. Public, dated, specific. Your existing agents already flag several of these: their output becomes a campaign feed, not a casualty. We add the pulls you do not have and act on all of it.
The two race each other. Four new campaigns every two weeks, each a permutation of buyer, signal and angle. Neither track gets protected. Whichever books qualified conversations gets doubled in the next cycle, and by month three the message that scales is proven, not guessed.
And LinkedIn runs where your buyers actually are. Regulatory affairs and quality leadership are professionally present on LinkedIn in a way few audiences are. Managed seats, human-paced, in a real voice, carrying the DoD origin story that cold email compresses too much.
Two managed seats, human-paced and proxied, pointed at regulatory and quality leadership in medical devices. Nothing needs to warm, so this channel produces conversations while the email infrastructure is still building.
FDA registration and clearance records, enforcement and recall data, MDR cohorts, plus the signals your own agents already track. Pulled fresh for each campaign and resolved to the named leader, never to an info@ inbox.
Signal campaigns and plain-offer campaigns run in parallel against the same goal, and neither gets protected. Whichever books conversations with device makers gets doubled in the next cycle.
Eight campaigns a month, each a permutation of buyer, signal and angle. By month three you own a ranked answer to which buyer converts best, and every SDR you hire afterward inherits it.
Every line here is a starting position, not a decision. The kickoff session exists so you can move these before anything is built, and Tuesday's call is where the arguing starts.
This table is the heart of the engagement, not the whole of it. The signal plays run next to straight-volume campaigns against the wider device-maker universe, and the two race each other. Nothing about Enginius needs repositioning. The platform needs to be put in front of the right regulatory, quality and engineering leaders with a reason to talk this week, and that is a campaign-count problem. The plays themselves are opening thinking: some ship as written, some change at kickoff, and some never go to market.
It is also the honest frame for the tier question in section 09. Eight campaigns a month works the device tracks properly. Sixteen runs life sciences and the engineering permutations in parallel rather than after. Either way, by month three you own a ranked answer to which buyer converts, and every SDR hire you make afterward starts with it.
The wider universe of US device makers, offered the platform plainly: first drafts of submission, change and quality documents from your own files, behind your firewall, deterministic and auditable. No trigger required, because every company in this universe is producing regulated documentation every single week.
Every sequence is three touches with one consistent ask: a fresh first email, a short threaded follow-up, then a fresh third angle, and then it stops. No breakup emails, no fake urgency, no "just bumping this." Sending runs on separate domains built for the campaign, never on enginius.ai, so your own domain and deliverability stay untouched.
Clearance records, published weekly, naming the company and the device that just made it through. Enforcement and recall data, marking the quality organizations whose documentation load just spiked. MDR cohorts, the CE-marked portfolios with technical files owed on a regulated clock. And your own agents' output: the leadership and regulatory changes your Claude-code trackers already flag, wired in as a feed instead of sitting in a channel nobody acts on.
How the data actually works, said plainly. A campaign pulls its records at the moment we build it, resolved to a named regulatory, quality or engineering leader with a verified contact. The campaign runs, we score it, and the plays that earn it get promoted to a standing pull so they keep feeding themselves. Everything is deduplicated and suppressed across plays, so no buyer hears from Enginius three different ways in the same fortnight.
Your team's agents represent real engineering investment and real market coverage, and the answer to "will we have to throw this away" is no, in writing. At kickoff we map what they track, integrate their output as campaign inputs, and build only what is missing. Where they are ahead of our own pulls, theirs win. The value we add sits after the signal: the list, the copy, the infrastructure, the send, and the conversation.
And the copy itself has to survive the most AI-suspicious readership in B2B. A quality director who spends her days rejecting unsupported claims will reject an email built from them in one glance. Every line is written to be read aloud, every claim traces to something you have actually said or shipped, and anything that smells generated dies in review. You approve the voice before the first send, and the sequences in section 05 are written so you can hold them to that standard right now.
Clearances are published weekly, in public, naming the company and the device. A regulatory team that just cleared knows precisely what the submission cost them in draft-assembly months, and the next submission is already forming on their roadmap. The congratulations is real, the question is one they have been asking themselves, and the 30-second first draft is a fact they will repeat internally whether they reply or not.
The widest universe of the five: established device makers whose products have been on market for years, where every engineering change drags impact assessments, file updates and regulatory review behind it. This is your own origin story pointed at its commercial twin, and it needs no signal at all, because in this universe change documentation is happening somewhere in the building every week.
Enforcement actions and recalls are public the moment they land, and they multiply exactly the documentation work your platform drafts: remediation plans, CAPA records, responses on a clock. This cohort gets found by the signal and written to with respect. The copy never recites what happened to them. It talks about documentation load the way a good peer would, and lets the reader make the connection privately.
Device makers selling into Europe owe technical documentation under MDR on a schedule nobody chose, with notified bodies backlogged and the same teams responsible for everything else. It is the rare why-now an entire cohort shares at once: dated, external, and nobody's fault, which makes it comfortable to say out loud in an email.
The DoD origin story is your best asset and the one cold email compresses too hard. On LinkedIn it can breathe: five years building under a Department of Defense contract, locked from commercializing until the work shipped, now open for exactly two verticals. This runs on the managed seats, human-paced, in a real voice, and it reads like a person because the sender is one. It also opens life sciences, the second phase-one vertical, without spending email volume on an unproven segment.
Plus the person who runs them. Which, on a team where half are rocket scientists and half are data scientists, is a hire that takes the wrong people off the platform that actually wins your deals.
Every tool above sits on our licenses and is run by our team. At the Engine tier you pay $4,000 a month and the stack behind it lists at more than that on its own, before anybody's time.
The working session: universe sized live on screen and cut by class and company size, the five buyers ranked, your agents' signal feeds mapped. Customers, board and partner relationships loaded as suppression. Both LinkedIn seats connected and the founder-story track goes out. Cold domains ordered and warming starts in parallel.
Clearance, enforcement, MDR and registry pulls built and resolved to named regulatory, quality and engineering leaders with verified contacts. Your agents' output connected as a campaign feed. First target lists back to you for review before anything sends.
All sequences written against the top two buyers and scored line by line, every product claim checked against what Enginius has actually said and shipped, in a voice you have approved. Low-volume soft launch on the new domains to prove deliverability before anything scales.
Cold plays running at full volume, replies routing to you same day. First two-week cycle scored and the next four campaigns built from what it showed. By the time your first SDR hire starts, the message they inherit has numbers behind it.
Our agentic systems listen to every US regulatory body continuously. When the FAA released a regulation affecting grounded-fleet predictive maintenance, the system found the affected operators, checked what each company's own site said about predictive maintenance, found the gap at GE Aerospace, resolved the right executive, and sent the message. The COO's reply asked how we got to it before he did. That deal closed for our client last week.
That is the exact mechanic this proposal points at medical devices: a regulator publishes, the affected cohort is identified the same day, and the message lands while the change is still news. Your own agents already do the listening half. Sections 02 through 05 are the other half.
Needed direct contact with decision-makers across thousands of US school districts, a universe that exists only inside public records, with the actual humans buried behind institutional names.
Mapped every administrator in every US public school district from public data, resolved them to verified direct contacts, and ran parallel campaigns off that dataset. That is the identical build to turning FDA registration, clearance and enforcement records into the named regulatory and quality leaders behind each device maker. It is the single most transferable thing in this list.
A saturated mid-market category, a sales team stretched thin, and a professional audience of accountants who dismiss generic outreach on sight. Regulatory and quality leaders read email the same way.
Intent-based outbound triggered on firms' own public activity, multi-touch across email and LinkedIn, with copy specific enough to survive an expert reader. The clearance and MDR triggers in your plays one and four are that exact mechanic pointed at regulated engineering instead.
Real credibility in the space but no systematic outbound, and no clarity on which of many possible angles would produce pipeline. Your question has the same shape: does the submission draft, the change package, or the MDR file open the door first?
40+ campaign types A/B tested weekly, doubling down only on what converted. This is the direct answer to the thing nobody can decide from a standing start: which buyer and which document type leads for Enginius? You do not have to pick in advance. Campaign velocity finds out with data instead of an opinion.
Owner-operators who do not answer generic email, in a category that closes on relationship, where the message had to reach the buyer at the moment the problem was live rather than whenever the campaign happened to send.
Signal data identified operators at the right moment, with sends timed to when those buyers were actually reachable. The finding that transfers to Enginius: a message that arrives inside the buyer's moment, a fresh clearance, a spiking document load, outperforms the same message sent cold. Your plays are built around exactly those moments.
Four campaigns every two weeks, eight a month. The medical device tracks worked properly, with your existing signal agents wired in from week one.
Eight campaigns every two weeks, sixteen a month. All five buyers plus the life sciences vertical run in parallel rather than sequenced.
Named, trained callers working your replies and dialing behind the email and LinkedIn touches. US and nearshore options, employer-of-record compliant, hired against the message the engine has already proven. Sized and priced per head once the first cycles show the reply volume.
| Onboarding & infrastructure setup | One-off | $1,000 |
| Total recurring |
Client: (fills from your signature) · Contact: (fills from your signature) · E-mail: (fills from your signature)
Selected Package: · Add-ons: none · Service Fees: , payable in advance per Section 7 · Billing Option: · Amount Due at Acceptance:
Onboarding Fee: , one time · Initial Service Term: months from kickoff, followed by month-to-month. Address and phone are captured on the onboarding form.
Services: Charm is a Go-To-Market Business Process Outsourcer (GTM BPO) providing Enginius.ai sales expertise and lead generation services per the selected package: lead acquisition against ICP criteria agreed at kickoff, systems and infrastructure setup, and campaign development with ongoing strategic support.
This Master Services Agreement ("Agreement") is entered into as of the acceptance date recorded on this page (the "Effective Date"), by and between Charm, registered as Didin Customer Service, LLC ("Charm"), located at 1220 E. Henry St, Tempe, Arizona 85281, and the client identified on the Order Form above ("Client").
Charm provides an AI-powered lead generation system with outbound efforts via email and LinkedIn campaigns promoting Client's goods and services for the purpose of generating and nurturing leads for Client (each, a "Campaign"). "Lead" means a potential customer contacted through LinkedIn or email for the purpose of Client offering its goods or services. Charm performs the services in a timely and workmanlike manner. Scope changes require written agreement before work begins, and Charm may charge reasonable costs associated with such changes.
Charm has full power and authority to enter this Agreement; performing it violates no other contract; lead sources infringe no third-party rights and promote no prohibited content; and performance complies with applicable law.
Client has full power and authority to enter this Agreement; performing it violates no other contract; and contact data Client furnishes has been diligently verified as serviceable and current.
All creative campaign assets are created by Charm. Charm may request existing content from Client to leverage in Campaigns. Client grants Charm a limited, non-exclusive, revocable, royalty-free license to use Client's marks solely to perform the services, ending with the Campaign or this Agreement. Client retains all rights in its intellectual property.
Before launch, Charm sends test materials for review. Client has a 24-hour window to object; silence is approval. Client may review each email before every new send, with the same 24-hour window.
The engagement runs an initial term stated on the Order Form (the "Initial Term"), then month-to-month. Fees for the Initial Term are committed at acceptance; neither Party may terminate for convenience during it. Either Party may terminate for material breach uncured within fifteen days of written notice. After the Initial Term, either Party may terminate on thirty days written notice. On termination, Client pays fees accrued through the effective date; if Client terminates for Charm's uncured material breach, prepaid fees for whole unstarted months are refunded.
All fees are payable in advance. The onboarding fee and first monthly period (or the discounted Initial Term fee, where paid-in-full is selected) are due at acceptance and presented for payment on this page through QuickBooks. Client authorizes payment by card or bank transfer through Intuit QuickBooks Payments; card and bank details are held by Intuit, never by this page. Subsequent monthly fees are invoiced in advance of each monthly anniversary of kickoff, invoice delivered seven days prior. Paid-in-full reflects the discount stated on the Order Form; the onboarding fee is not discounted. Late amounts incur 1.5 percent per fourteen days past due, and the program pauses seven days after written notice of nonpayment.
Charm retains access to Client data to perform the services. Client may view and export all Campaign data at its discretion.
Each Party protects the other's Confidential Information with at least commercially reasonable care, uses it only to perform this Agreement, and limits disclosure to those under equivalent obligations. Standard exclusions apply (public information, prior knowledge, independent development, rightful third-party receipt, Campaign materials and metrics). Trade secrets are held indefinitely; other Confidential Information for five years. Legally compelled disclosure requires reasonable prior notice.
Charm shall not use Leads provided by Client for the benefit of competing ventures, during the term and for one year after.
Each Party indemnifies the other against losses, including reasonable attorneys' fees, arising from its breach of this Agreement.
Arizona law governs; exclusive jurisdiction lies in Arizona courts.
Invalid provisions do not void the remainder; the Parties negotiate replacements in good faith preserving original intent.
No assignment without written consent, except to a merger successor, asset purchaser, or commonly controlled entity.
Charm performs as an independent contractor. No partnership, joint venture, agency, or employment is created.
Both Parties comply with all applicable Do Not Call and Do Not Contact regulations, maintain lists against registries, honor DNC requests promptly, and keep required records.
No waiver except in signed writing. Remedies here are in addition to those at law or equity.
Acceptance through this page, together with the typed name and the recorded SHA-256 hash of the Order Form and this Agreement as displayed, constitutes a binding electronic signature under the U.S. ESIGN Act and UETA. Charm: Didin Customer Service, LLC, by Chris Booth, Owner.
Ten minutes. Brand voice, ICP, suppression, access.
2 · Book your kickoff →The onboarding discovery session.
Arrives with your kickoff confirmation.
If the engagement has not returned its cost by the end of month three, we run month four entirely at our cost, full effort, nothing held back, and we will connect you with people we have run that month for so you can hear how it went. And at month three you choose: keep going, or take the campaign matrix, the copy and the target pulls and run them yourself, with the SDRs you hire inheriting all of it. They are yours either way.
Claim your guarantee →The working session where we size the device-maker universe live, rank the five buyers, map what your agents already track against what we pull, and load your customers, board and partners as suppression. It ends with a target list on screen, not with a follow-up email.
The founder-story track goes live on LinkedIn in the first few days while domains warm in parallel. The clearance, enforcement and MDR pulls get built, your agents' feeds get wired in, and you review every target list before a single message sends.
Cold plays live around week four at full volume. Every two weeks a fresh cycle of four campaigns ships, built from what the last cycle showed. When the numbers say which message converts, your SDR hires start with a proven script instead of a guess. At month three, you choose what happens next.
Pick a kickoff date. Week one is the parameter session, the ranked buyers, your agents wired in as a feed, the suppression list loaded, and the founder story already reaching regulatory leaders on LinkedIn. None of that waits on infrastructure to warm.
Pick your kickoff date →