Drop in your data,
your AI uses it instantly.
AI-native PostgreSQL you call straight from Claude Code / Cursor / Codex. HTAP + pgvector + multi-tenant + private networking. Pre-aggregate server-side and cut AI token spend up to 70%.
Sound familiar?
Four walls every team hits when wiring AI / LLM into their product.
Data is scattered, AI can't see across it
DBs, file stores, SaaS tools, and internal docs are spread out — you write yet another ETL just to connect them. Hundreds of hours/month vanish into pre-processing for AI.
Security implementation eats months
Encryption, MFA, audit logs, backups, WAF… building each yourself takes 6 months. But customers want access tomorrow.
GDPR / privacy laws are a constant headache
PII detection, masking, residency, cross-border consent… consulting legal every step grinds the product to a halt.
You want Claude / Cursor to talk to your DB directly
It's unclear which DBs actually speak MCP. You end up writing your own API plus auth and rate limits — a month of work that holds your customers back.
Drop the data in. We do the rest, automatically.
Six months of plumbing, condensed into one DB. The only code you write is INSERT.
All your data in one place, automatically AI-ready
Reach the same data from your app and from AI via REST, SDK, SQL, and MCP. Vector search included via pgvector — one database. You just INSERT.
- MCP-native: connect directly from Claude / Cursor
- Vector + hybrid search via pgvector
- Transactions + analytics + vector search in one foundation
- Auto-ingest connectors for external DBs and SaaS
Zero security work. Production-quality from day one.
2FA, Cloud Armor WAF, encryption, audit logs, automated backups, prompt injection defense, intrusion detection — every line item enterprise procurement asks for, already inside. SOC2 / ISO27001 audits become a formality.
- TOTP 2FA (Google Authenticator / 1Password)
- Cloud Armor + Adaptive Protection (autonomous DDoS defense)
- Immutable backups (ransomware-proof, 7-day retention lock)
- Prompt injection guard + AI output leak redaction
Governance via RBAC, audit logs, tenant isolation, and private networking
Cloud DLP API + industry dictionaries auto-detect PII and apply k-anonymization. SSO, VPC Service Controls, CMEK, and residency choice clear enterprise procurement on the first pass. The legal-review back-and-forth disappears.
- PII / national ID / sensitive data — auto-detect + anonymize
- GDPR / APPI / CCPA logs + cross-border consent management
- SSO (SAML/OIDC) + VPC + CMEK (Enterprise)
- Residency: asia-northeast1 / europe-west1 / us-central1
What you get back
What choosing NW DB returns to your engineering team.
Simple Pricing
Free trial included on every self-serve plan: Hobby 3 days, Pro & Team 7 days, Scale 14 days. Commercial use is completely free.
Hobby
For solo developers, startups, and MCP-first workloads
- Database 2GB included
- MCP / REST / GraphQL / SQL
- Realtime subscriptions
- MCP endpoint + REST/SDK
- 100K API calls / month
- 1M AI tokens / month
- Vector search 10K queries
- First-class Claude Code / Cursor / Continue support
- File storage 500MB
- Google sign-in + 2FA
- Row Level Security
- Commercial use free
Pro
For growing products, small businesses, and mid-to-large data workloads
- All Hobby features + 5× quotas
- 500K API calls / month
- 10GB Storage
- 1M AI tokens / month (overage ¥100 / 10K tokens)
- 50K vector queries
- 5GB file storage
- 50K vector queries/month
- Email support (24h response)
Team
For production workloads, multi-user teams, and HA requirements
- All Pro features + warm-standby HA
- 1.5M API calls / month
- 30GB Storage
- 3M AI tokens / month (overage ¥100 / 10K tokens)
- 2 team seats included
- Read replica + connection pooling
- Read replica + connection pooling
- Email support (12h response)
Scale
Mission-critical commercial workloads with SLA 99.9% guarantee
- All Team features + hot standby + auto-failover
- SLA 99.9% guarantee
- 5M API calls / month
- 100GB Storage
- 50M AI tokens / month
- Point-in-Time Recovery (PITR)
- 10 team seats included
- Point-in-Time Recovery (PITR)
- Priority support
- Priority support
Enterprise
For organizations with advanced security and compliance needs
- Dedicated environment
- SLA with uptime guarantee
- Instant response + refund policy
- Dedicated support team
- CMEK encryption
- VPC private networking
- SSO (SAML / OIDC)
- Audit log retention
- Analytics export integration
- Dedicated setup + custom SLA
Simple Architecture
Built on proven open-source foundations. Deployed on Google Cloud.
High-performance database layer
Auto-generated REST / GraphQL / vector search. Transactions, analytics, and AI integration in one foundation.
MCP / REST / GraphQL Gateway
Connects to Claude / Codex / Cursor / Continue. Vendor-neutral API layer with MCP endpoint, auth, rate limiting, and WAF
AI Inference + Vector Search
Heavy aggregation via a DuckDB sidecar, vector search via pgvector, optional BigQuery.
Let's start with a consultation
Hobby / Pro / Team / Scale are self-serve and instantly available. For Enterprise, a 30-minute Discovery Call covers your use case and the right plan.
Request consultation