Build vertical AI products with platform discipline.

SoloFrame powers the flagship shared-engine contrast: DWA and GTM-OS. BizFrameHub, Colombian Spanish Academy, and 60-Day Founder extend the SVTech proof into SMB AI education, localized language learning, and founder go-to-market execution.

Agentic AI - not a claim, a result

The architecture is why this was possible. One founder. Multiple proof surfaces.

Most agentic AI implementations fail at the quality bar: coherence collapse, hallucination at scale, no ground-truth anchor. The manifest-driven architecture is what made the clinical, commercial, SMB, and localized learning outputs coherent enough to ship.

6
proof surfaces
922
clinically-validated lessons
57
BizFrameHub courses
4d
localized learning build
Why most agentic AI fails
  • Coherence collapse across long content chains
  • Hallucination accumulates without a ground-truth anchor
  • No promotion gate - plausible-looking output ships
  • Quality degrades without enforced structural guardrails
What made this work
  • Manifest guardrails at engine level - constraints enforced architecturally, not by prompt
  • Evidence grading as anchor - every lesson cites primary literature
  • PMHNP validation loop - board-certified clinicians as ground-truth signal
  • MAIA promotion gate - models rejected when they miss the bar, documented when they do

Phase 2 in DWA still requires zero engine changes. BizFrameHub is launching today as a 57-course AI Business Academy for SMBs and professionals from the websites workspace. Colombian Spanish Academy proves a localized education/community vertical can move in days. Competitors cannot close that gap by hiring alone. The operating discipline is the advantage.

See the DWA vertical deep-dive ->
The architecture in one frame

Flagship contrast plus expansion proofs. One discipline, many outputs.

DWA and GTM-OS remain the cleanest architecture proof: opposite compliance posture, same engine. BizFrameHub and Colombian Spanish Academy add the generalization proof: SVTech can package new domains into credible product surfaces without turning each project into a sprawling custom build.

● Live now · 4th vertical
60-Day Founder - AI Startup Academy
Startup AI

Now live on Mono-PaaS: an AI-native startup academy - 257 lessons, 6 tracks, an AI build coach, and a builder certification. Proof the same engine spins up new schools, audiences, and business lines through configuration, not rebuilds. See the vertical →

What will you build on Mono-PaaS?
The next vertical starts here
Expansion ready

Build your new academy, operator program, or branded AI learning business on the same manifest-driven platform already powering multiple live schools.

Layer 2 · Adapters1 @svtech/* package · 10 wired in production · 3 aspirational
@svtech/adapter-nodebbr2-storageresendpgvectorbrevoattiopipedrivehunternotionwhatsappbadgrflarumpolarstripe
Layer 1 · Engine packages · 18 @svtech/* packages12 wired into both flagships + 4 selective
@svtech/auth
@svtech/coaching
@svtech/content
@svtech/contracts
@svtech/db
@svtech/logging
@svtech/manifest
@svtech/onboarding
@svtech/presence
@svtech/profile
@svtech/redis
@svtech/safety
@svtech/llm
@svtech/voice
@svtech/security
@svtech/http
wired into both flagship wiring.ts files
selective (voice on demand, http light)
Deployment substrate · single Dockerfile, multi-node by design
Dokploy
Control plane · GitHub-App webhook · autoDeploy on main
Traefik
Ingress + TLS · Let's Encrypt · HTTP→HTTPS upgrade
Docker Swarm
Multi-node scheduling · stateless replicas
Postgres · Redis · R2
State layer · pgvector for GTM, JSONB+cosine for DWA

Hard rules: no vertical may fork the engine; nothing lands in packages/ until the abstraction is earned by multiple products or a committed next use case.Full architecture →

Why the engine works

Three properties the flagship products and newer proof builds share, each a direct consequence of the design decisions made from day one - and each translating 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; later proof builds can move in 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 as a v2 target. v1 ships audited application-level scoping; v2 moves the boundary into Postgres row-level security keyed on a per-request tenant GUC. The tenantLeakHarness ships with that rollout 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",
    "storage": "r2",
    "mail": "resend"
  }
}

Six proof surfaces, one company thesis.

DWA and GTM-OS are the flagship engine contrast: PHI vs non-PHI, MAIA vs noop classifier, same shared core. BizFrameHub and Colombian Spanish Academy prove SVTech can turn domain knowledge into new commercial and localized learning surfaces quickly.

2
Flagship verticals
2
Expansion proofs
1
Platform engine
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.

Why now

The education market is enormous. The wedge is fast-deployable AI-native vertical academies.

SVTech is not trying to be another LMS. The sharper claim is school-in-days deployment: your branded academy, live in days, with an AI coach that knows your curriculum, and a safety layer you can dial up to clinical grade.

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

Inference runs CPU-only for serving-path isolation (50-200ms target latency, no GPU contention with application workload). Training and retraining are GPU-bound; the data-flywheel architecture unlocks once live-customer signal accumulates and dedicated GPU capacity is available.

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 an applied-AI company. It builds SoloFrame - a multi-tenant vertical PaaS - and operates a growing portfolio of vertical AI products and proof surfaces: Digital Wellness Academy, GTM-OS / SoloFrameHub, BizFrameHub, Colombian Spanish Academy, and ContentForge (internal content 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 uses SoloFrame as the platform layer behind multiple products and proof builds; the engine is shared, not forked.
What vertical SaaS are running on SoloFrame today?+
The flagship contrast is Digital Wellness Academy and GTM-OS / SoloFrameHub: DWA is a HIPAA-aware mental-health education platform with 77 courses, 922 lessons, 21 clinical assessments, and MAIA-served safety; GTM-OS is a non-PHI founder go-to-market operating system with a 49-course / 974-lesson curriculum, DISC roleplay, execution AI, and integrations. The newer expansion proofs are BizFrameHub, a 57-course AI Business Academy for SMBs and professionals, and Colombian Spanish Academy, a rapid localized learning vertical with 20 courses across 6 tracks plus a Bogotá community guide module.
How does SoloFrame ensure tenant isolation?+
v1 enforces tenant scoping in application code with audited query paths. v2 moves isolation into Postgres row-level security keyed on a per-request tenant GUC (Grand Unified Configuration variable), with two Postgres roles (neither with BYPASSRLS) and a withTenant(ctx, fn) contract every DB-touching engine must obey. A tenantLeakHarness lands with the v2 rollout 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 the platform's classifier service - a CPU FastAPI sidecar running on every user-generated message inside HIPAA-aware SoloFrame verticals, before any LLM sees the text. The current production model is sentinet/suicidality (ELECTRA-base, ~110M params, CC0 baseline, C-SSRS-trained). Validated F1 0.93, crisis recall 1.000 (6/6), trap-FP 0 (0/2) on the May 2026 harness. Inference at 22.3 ms/sample on commodity CPU; ~$0.0001 per inference. The 5-stage retrain pipeline (services/maia/MODEL_CARD.md) promotes a successor only when it measurably beats the sentinet baseline on held-out validation. Audit-trailed by per-inference model_version. failSafe="closed" routing means a MAIA outage produces a conservative 988-surfaced response, never an unguarded coaching call.
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. The build itself runs the same way: a self-hosted Huly issue tracker wired to Claude over MCP, with GitHub pull requests syncing in as tracked issues and a deployed agent triaging them in chat. 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 ->
New on Digital Wellness Academy · June 1, 2026
Platform velocity proof

Neuroplasticity School (Mind & Performance) ships on schedule

A new evidence-graded school added to DWA on the existing engine - manifest-driven, no routing or engine changes. 17 courses · ~225 lessons across five pillars, each lesson carrying an explicit evidence grade (Strong / Moderate / Emerging) and embedded into the same RAG corpus MAIA uses for "why does this work?" coaching. It is the Mono-PaaS thesis in one shipment: a new school drops in as content, not a fork.

Brain Science FoundationsMindset & Cognitive RewiringMindfulness & AttentionBody–Brain PerformanceIdentity & Purpose
Read the whitepaper →DWA proof packet →
Just shipped · New book & academy

Announcing 60-Day Founder - the book and the academy.

A new book, 60-Day Founder: A Builder's Guide to Production AI, and the AI-native academy built on it - now live as the fourth production vertical on the SoloFrame engine. Learn → Do → Measure → Refine: idea to VC-ready evidence in sixty days, with an AI build coach reading your live build context.

The academy · live
60-Day Founder

257 lessons · 32 courses · 6 tracks. AI build coach, 8 artifact workshops, founder-readiness assessment, and the Certified Builder credential.

Explore the academy →
60-Day Founder: The Solo Builder's Guide to Production AI - book cover
The book · 2026 edition
A Builder's Guide to Production AI

13 chapters · 72,000+ words · 23 figures. Foundations of AI entrepreneurship, then the day-by-day 60-day execution roadmap.

Read the book →

Fourth vertical, same engine - shipped as manifest-declared content, not a fork. Live at mono-startup.soloframehub.com.