SVTech Consulting · SoloFrame

A vertical PaaS for AI-native healthcare and revenue products

SVTech is building SoloFrame - a multi-tenant vertical PaaS with compliance, tenant isolation, and AI orchestration built in. We're extracting it from DWA (HIPAA-aware healthcare) and GTM-OS (revenue operations) - the two vertical SaaS entering beta on it today - and the platform is designed to scale from a single node to a multi-server deployment without a rewrite.

Two vertical SaaS. One vertical PaaS in extraction.

Why SoloFrame is the engine we're extracting

Three properties of the architecture we're pulling out of the two verticals. Each one is a direct consequence of the design decisions made from day one - and each translates into how fast future verticals ship, how safely we operate, and how the portfolio compounds.

Speed

Launch a new vertical in weeks, not quarters

A vertical is defined as configuration - not code. New products compose from shared primitives instead of duplicating systems. The first flagship took months; the second took days.

Safety

Compliance as a setting, not a rewrite

Healthcare verticals set phi: true and inherit HIPAA-aware behavior. Non-healthcare verticals get none of that overhead. The manifest wires up retention, guardrails, and classifier sidecars; the engine enforces it.

Compounding

Every vertical sharpens the next

Shared classifiers, shared assessments engine, shared analytics pipeline. The safety model a healthcare vertical trains today improves every future clinical deployment. Longitudinal data compounds into a moat competitors can't buy.

Vertical AI portfolios fail in three predictable ways

Every studio we've watched try to build multiple AI products has hit the same three walls. SoloFrame was designed against each of them from the beginning, because we're the studio too.

1
Fork hell

Each new vertical becomes a new codebase

By the third product, you're maintaining three stacks - each with its own auth, tenancy, billing, and AI wiring. The platform vision collapses into backlog.

SoloFrame's approach

Manifest-driven verticals. Each product is configuration, not a forked codebase. One engine, many outputs. An admission rule keeps the core small: no package lands without ≥2 consumers.

2
Tenant leakage

Application-code isolation is a breach waiting to happen

When isolation lives in WHERE clauses and middleware instead of the database, one missed condition in one query can expose another tenant's data.

SoloFrame's approach

Database-enforced isolation. Postgres row-level security at the data boundary, keyed on a per-request tenant GUC. A tenantLeakHarness runs in CI and blocks the build on any regression.

3
Generic AI safety

Off-the-shelf moderation misses domain-specific risk

The signals a healthcare deployment needs to catch are not the signals a forum moderator needs to catch. LLM-provider safety filters were trained to catch different things.

SoloFrame's approach

Pluggable safety layers. Verticals that need a purpose-trained classifier attach one (DWA's MAIA); verticals that don't get none of the overhead. Wired in by manifest flag, not app code.

THE MANIFEST

A vertical is configuration, not a codebase

A typed JSON manifest describes everything the engine needs to know about a vertical: compliance posture, AI model choices, adapter wiring, assessments, branding, roles, billing plans. Twelve primitives. All validated. All version-controlled.

Changing a vertical means editing configuration. Adding a vertical means composing primitives. Neither requires changing the engine.

manifest.json (excerpt)
{
  "id": "dwa",
  "compliance": {
    "phi": true,
    "retentionDays": 2557,
    "guardrails": [
      "never-diagnose",
      "always-988-on-crisis"
    ]
  },
  "ai": {
    "classifier": "maia",
    "coachingModel":
      "claude-haiku-4-5"
  },
  "adapters": {
    "forum": "flarum",
    "payments": "polar"
  }
}

Two vertical SaaS built. One vertical PaaS in extraction.

DWA and GTM-OS shipped first as standalone vertical SaaS products. SoloFrame is the shared spine we're now pulling out of them - identity, tenancy, AI orchestration, adapters, manifests. As the extraction completes, both verticals run on the same vertical PaaS; neither is a fork, because there's nothing to fork yet.

2
Vertical SaaS built
1
Vertical PaaS in extraction
0
Forks (admission rule)
The market position

An unclaimed middle between two well-served markets

Foundation-model providers ship models, not products. Vertical SaaS companies ship products, not platforms. AI-native point products ship one use case. The middle - a platform that builds and operates vertical SaaS with a shared AI-native core - is largely unclaimed.

Foundation model providers
Models, not products

OpenAI, Anthropic, Google, Mistral. The infrastructure every AI product runs on - but none of them ship vertical SaaS, and none have visibility into a vertical's data flywheel.

Vertical SaaS shops
Products, not platforms

Veeva, Toast, Procore and the rest. Excellent at one industry vertical. Rarely AI-native at the core. No reusable spine for the next vertical; each new category is a fresh raise.

SVTech · the unclaimed middle
Platform + verticals + flywheel

A vertical PaaS that builds and operates its own vertical SaaS. Each vertical's real-world usage feeds the shared data flywheel - sharpening the platform that powers every future vertical.

Foundation model APIs
Serve every AI product. Build no verticals.
AI-native point products
One vertical, one product. No platform thesis.
Traditional vertical SaaS
Industry-specific products. Rarely AI-native.
SVTech
Vertical PaaS + operated vertical SaaS. Flywheel enabled.
ARCHITECTURE DISCIPLINE

Modular monolith. Horizontally deployable. Redundant by design.

SoloFrame is being built as a horizontally deployable modular monolith - one codebase, stateless app replicas, shared state in Postgres / Redis / object storage, tenant routing resolved at request time. It runs comfortably on one node today and scales to a multi-node deployment under Dokploy + Traefik + Docker Swarm as workloads grow. Multi-node isn't just for scale - it's redundancy: stateless replicas behind Traefik mean a failed node doesn't take the platform down, and rolling deploys don't require maintenance windows.

Pooled shared runtime
Default tier

One codebase, multi-tenant, stateless replicas behind Traefik. Today's beta footprint fits on a single node.

Isolated topology
Licensed & premium

Dedicated app replicas, dedicated DB, dedicated storage - a deployment topology choice, not a code fork.

GPU-bound training
Off the serving path

Model retraining (MAIA, ClinicalBERT, sentiment, risk) runs on GPU instances out-of-band. Inference on the safety path stays CPU-bound for reliability.

Stack: Dokploy control plane · Traefik ingress · Docker Swarm multi-node scheduling · Postgres (pgvector) · Redis · S3-compatible object storage. No Kubernetes, no microservices.
STRUCTURAL COST EFFICIENCY

Root-server deployment, not Vercel-style per-invocation pricing

We deploy to VPS and root-server infrastructure via Dokploy, not to managed-PaaS hosts that bill per request, per function invocation, or per bandwidth unit. That matters for AI-heavy workloads: a single thoughtful LLM session can fire dozens of inference calls and streaming tokens, and per-invocation pricing converts engagement into variable cost that fights margin.

Running on root servers gives us predictable capacity at a predictable price. The modular-monolith-on-Swarm pattern adds redundancy and horizontal scale by adding nodes - not by paying a hosting vendor more per user. That structural advantage compounds as usage grows.

FREQUENTLY ASKED

Questions we hear a lot

Fast answers to the things clients, partners, and investors most often ask.

What is SVTech Consulting?+
SVTech Consulting Services LLC is a Wyoming applied-AI company. It builds SoloFrame - a multi-tenant vertical PaaS - and operates its two first-party vertical SaaS, Digital Wellness Academy (healthcare) and GTM-OS (revenue operations). Founded by Mike Sullivan in December 2022.
What is SoloFrame?+
SoloFrame is SVTech's vertical PaaS - a modular-monolith, multi-tenant platform for AI-native vertical SaaS. Verticals resolve as content plus a typed manifest, not branching codebases. It is designed for horizontal deployment under Dokploy + Traefik + Docker Swarm, with stateless app replicas and shared state in Postgres / Redis / object storage.
What is a vertical PaaS, and how is it different from a foundation-model provider or a vertical SaaS company?+
Foundation-model providers (OpenAI, Anthropic, Google) ship models, not products. Vertical SaaS companies (Veeva, Toast, Procore) ship products, not platforms. A vertical PaaS sits between them: it builds and operates vertical SaaS on a shared AI-native core, with verticals feeding a data flywheel that sharpens the platform. SVTech runs two vertical SaaS on SoloFrame today, with the PaaS under active extraction from both.
What vertical SaaS are running on SoloFrame today?+
Two, both entering beta: Digital Wellness Academy - a HIPAA-aware mental-health education platform that psychiatric practices license to extend care between sessions, with 43 courses and 592 lessons across therapeutic and optimization tracks - and GTM-OS - a B2B learning platform for tech founders and teams with DISC-based peer pod matching, AI coaching, and sales-practice roleplay.
How does SoloFrame ensure tenant isolation?+
Shared Postgres with row-level security keyed on a per-request tenant GUC (Grand Unified Configuration variable). Two Postgres roles, neither with BYPASSRLS. Every DB-touching engine obeys a withTenant(ctx, fn) contract. A tenantLeakHarness runs in CI and blocks the build on regression. Dedicated-DB deployment is available as a paid SKU for licensees who require it.
What is MAIA, the distress classifier?+
MAIA is a fine-tuned DistilBERT model (66M parameters) that runs as a CPU-only FastAPI sidecar on every user-generated message inside HIPAA-aware SoloFrame verticals - before any LLM sees the text. It produces a three-tier output (None / Mild / Crisis) with zero-knowledge at inference: the input text is never stored or logged. Crisis-level classifications trigger 988 Lifeline escalation and provider notification. MAIA is continuously retrained on accumulating platform analytics.
How does SVTech engage with clients?+
Three stages. Stage 1 - Custom Vertical (active now): SVTech designs, builds, and operates a fully-branded vertical SaaS on SoloFrame, starting from a paying pilot customer's real pain point. Stage 2 - Licensed (2026-2027): partner organizations license an existing SVTech vertical, re-skinned for their brand, in a dedicated environment. Stage 3 - Studio (2027+): self-serve composition from a locked allowlist, only after the engine is battle-tested.
Why a monolith instead of microservices?+
A modular monolith gives founder-scale operational simplicity while remaining horizontally deployable. Shopify, Basecamp, and Amazon Prime Video have each publicly defended this pattern for the same reasons: faster shipping, coherent debugging, one identity and one tenancy model across all verticals, and scale through added replicas rather than service proliferation. SoloFrame is stateless at the app layer and shares state in Postgres / Redis / object storage, so multi-server deployment is a topology choice, not a rewrite.

Three reasons to reach out

A custom vertical built for your organization. A licensed deployment of an existing one. An investor conversation. Pick the one that fits.

Contact SVTech ->