Govern Your AI.
Protect Your Data.
Control Your Spend.
The first enterprise AI governance platform that combines cross-provider gateway, real-time budget enforcement, and policy-driven compliance — all in one platform.
Live demo with seeded data — no signup, no installation.
The AI Spend Crisis
Shadow AI Spend
70% of AI projects fail to reach production. Unmanaged spend across departments with no centralized visibility.
Compliance Risk
EU AI Act enforcement begins Aug 2026. Penalties up to €35M or 7% of global revenue. Companies have 12–18 months to comply.
One Platform. Total AI Governance.
A proxy gateway between your organization and every AI provider — with real-time control, visibility, and compliance.
Visibility
See exactly who is using AI, how much it costs, and what data is being sent — across all providers, in real time.
Control
Enforce budgets at any granularity (org → team → role → entity). Block risky requests before they execute.
Compliance
Meet SOC2, GDPR, and EU AI Act requirements with audit trails, prompt scanning, and data residency controls.
Flexibility
Support any AI provider (OpenAI, Anthropic, Local Models) and any deployment model (SaaS, Self-Hosted, Air-Gapped).
Every request passes the gates before it leaves your perimeter
Request pipeline — 12 steps, ~20–50ms overhead
pre-execution enforcementMarket Opportunity
42.3% CAGR through 2030 (Grand View Research)
Enterprises spending >$10K/mo on LLM APIs
2% of SAM via PLG + Enterprise motion
Market Growth Drivers
- LLM API spend: $100M (2023) → $2.5B+ (2025) — 25x in two years
- 88% of organizations use multiple AI providers
- CEOs plan to double AI spending in 2026
- EU AI Act enforcement wave starting Aug 2026
Target Customer
- Mid-market to enterprise (50–5,000+ employees)
- Using multiple AI providers
- Need visibility, control, and compliance
- Beachhead: Technology/SaaS, then Financial Services, Healthcare, Government
Core Platform Architecture
AI Access Gateway
- 12-step request pipeline with ~20–50ms overhead
- Auth → Prompt Scan → Policy Eval → Budget Check → Forward
- Supports OpenAI, Anthropic, Google, AWS, Azure, Cohere, Local Models
- Hybrid key model: managed or customer-provided keys
- 10,000+ requests/second target
Management Plane
- 4-persona dashboards (Admin, Manager, User, Agent Owner)
- OPA/Rego policy engine with 20+ pre-built templates
- Real-time budget enforcement (soft alert → hard block)
- Tiered approval engine (auto / manager / escalation)
- 10 alert types across Email, Slack, Teams, PagerDuty
Go (fasthttp)
PostgreSQL + TimescaleDB
Redis + Kafka
Competitive Advantage
No one offers all three: cross-provider governance + enterprise compliance + budgeting.
| Feature | Quorix | Portkey | Helicone | LangSmith | Braintrust |
|---|---|---|---|---|---|
| Cross-Provider Gateway | ✓ | ✓ | ✓ | — | — |
| Budget Enforcement | ✓ | — | — | — | — |
| Policy Engine (OPA/Rego) | ✓ | — | — | — | — |
| Prompt Auditing (PII) | ✓ | — | Partial | — | — |
| Self-Service Approvals | ✓ | — | — | — | — |
| Enterprise SSO (SAML) | ✓ | — | — | ✓ | — |
| SOC2 Compliance | ✓ | — | — | ✓ | — |
| Self-Hosted Option | ✓ | — | — | — | — |
| Shadow Spend Discovery | ✓ | — | — | — | — |
Scroll →
Portkey was acquired by Palo Alto Networks (May 2026) — validating enterprise demand for AI governance. But their product lacks budget enforcement, policy engine, and SSO.
5 Unique Differentiators
Visual Policy Engine
Industry-standard OPA/Rego with drag-and-drop builder. Once policies are written, switching is extremely painful — deep technical lock-in.
Prompt Auditing
Inline PII scanning of input AND output before forwarding. Progressive: regex → Presidio → ML-based. No competitor offers this.
Self-Hosted Deployment
SaaS, Private Cloud, Customer Cloud, On-Premise, Air-Gapped. No AI-native tool offers this flexibility.
Shadow Spend Discovery
Find unmanaged API keys via invoice matching or network probe. Free audit demonstrates immediate ROI for CFOs.
HRIS/ITSM Integration
Embed into existing workflows: Workday, BambooHR, ServiceNow, Jira. Deep organizational embedding creates switching costs.
Business Model
- 1M requests/month
- 3 API keys
- 5 team members
- Customer-provided keys
- Basic dashboards
- 500K requests/month
- SSO (Okta/Azure AD)
- Budget enforcement
- Advanced dashboards
- Team management
- 5M requests/month
- OPA/Rego policy engine
- Prompt auditing
- Compliance reporting
- Unlimited team
- Self-hosted option
- Custom integrations
- Dedicated support
- EU AI Act compliance
- SLA guarantees
Alternative: 5–10% of managed AI spend (minimum $1,000/mo)
Land & Expand Motion
Expansion Triggers
- Usage gates (exceeded request limits)
- Feature gates (need SSO, policy engine)
- Team growth (5+ → 20+ → 100+ members)
- Compliance needs (SOC2, EU AI Act)
- Value realization (first budget alert, first PII detection)
Viral Loop
- Each user invites 2–3 colleagues (target: 2.5x viral coefficient)
- Free tier → Pro → Business → Enterprise
- Developer-led bottom-up adoption
- CTO/CFO-led expansion
Revenue Trajectory
20 Enterprise + PLG mid-market
50 Enterprise + 500 mid-market
100 Enterprise + 1,500 mid-market
200 Enterprise + 3,000 mid-market
350 Enterprise + 5,000 mid-market
Unit Economics
- Blended LTV:CAC: 4:1 (base case)
- CAC payback: < 12 months by Year 3
- 120% Net Revenue Retention
- 75% gross margin by Year 3
Cost Structure
- Pre-Seed: $80K/mo burn, 5–8 team
- Seed: $150K/mo burn, 15–20 team
- Series A: $300K/mo burn, 40–50 team
- Engineering: 40–50% of burn
Key Milestones
- Month 6: 2:1 LTV:CAC
- Month 12: 3:1 LTV:CAC
- Month 18: CAC payback < 12mo
- Month 36: Blended LTV:CAC > 5:1
Go-To-Market Strategy
Wave 1: Design Partners
- Early-stage startups, dev shops, AI labs
- Personal founder outreach (LinkedIn, email)
- Hacker News, Reddit, Twitter
- Free access for 6 months + co-development
Wave 2: PLG Customers
- Mid-market tech companies (50–500 employees)
- Content marketing, SEO, Product Hunt
- Free tier → Pro upgrade ($99/mo)
- Conference presence, partner integrations
Wave 3: Enterprise
- Fortune 1000 companies (1,000+ employees)
- Enterprise sales team (3–5 AEs)
- Case studies, EU AI Act messaging
- Channel partners (MSPs, SIs)
Who We Serve
10 personas driving product decisions and GTM strategy
Primary Beachhead
Technology / SaaS — CTO, VP Engineering, Head of AI/ML. ACV: $30K–$60K. PLG-first, bottom-up adoption.
Expansion Markets
Financial Services (ACV: $100K–$250K), Healthcare (ACV: $100K–$200K), Government (ACV: $200K–$500K)
Regulatory Tailwinds: EU AI Act
The compliance window is closing — companies have 12–18 months before the main enforcement wave.
Manipulation, social scoring, emotion recognition
Foundation model providers must comply
Chatbot disclosure, AI content labeling, deepfake marking
Biometrics, finance, healthcare, justice
All provisions active
Penalties
- General violations: up to €15M or 3% of global turnover
- High-risk violations: up to €35M or 7% of global turnover
- Incorrect information: up to €5M or 5% of global turnover
How We Address It
- Risk assessment → Policy Engine (OPA/Rego)
- Transparency → prompt auditing, usage logging
- Human oversight → approval workflows
- Data governance → PII scanning, data residency
- Monitoring → dashboards, alerting, data export
Implementation Roadmap
8 senior engineers, 4 parallel workstreams, 15 months to full platform
Phase 0: Demo
- Functional demo
- Design partner acquisition
- Architecture validation
Phase 1: MVP
- Gateway core (Go + fasthttp)
- Budget enforcement
- SSO (Okta SAML)
- OPA/Rego policy engine
- Prompt scanning (regex)
Phase 2: Policy
- Visual policy builder
- Presidio PII scanning
- Shadow Spend Discovery
- Hard budget blocking
Phase 3: Enterprise
- SOC2 compliance
- Self-hosted (Docker)
- HRIS integration
- Cost per output metrics
Phase 4: Scale
Kubernetes self-hosted · Provider plugin system (codeless onboarding) · ML-based prompt scanning · Global edge deployment
Team & The Ask
Team Composition
The Ask
- 18 months runway at $80K/mo (Pre-Seed)
- 8 senior engineers across 4 disciplines
- Phase 0 Demo in 6 weeks
- MVP with first paying customers by Month 4
- 25 design partners in first 90 days
See it working on live data
Log in to the demo — seeded providers, keys, budgets and usage — and walk through the platform in minutes.
Take the tour →Demo credentials are filled in for you.