for enterprise

ai coding your security team approves.

sso, scim, rbac, audit.db, vpc / on-device models, guest access. an ai ide that procurement actually says yes to.

saml sso, rbac, scim

okta, azure ad, google workspace. role-based access at every layer. automated provisioning and de-provisioning.

local audit.db

every tool call, every model request, written to a portable sqlite. queryable by your siem, owned by you.

vpc & on-device llms

point hellocode at your own llama / qwen / claude-on-bedrock deployment. code never leaves your perimeter.

guest access

scoped, time-bound seats for contractors and reviewers. revoke in one click, with an audit trail.

compliance docs

dpa, sub-processor list, security whitepaper. compliance certifications in progress, ready before the security questionnaire shows up.

dedicated support

named cs engineer, slack channel, 24h sla on critical issues. quarterly business reviews.

procurement faq

where does our code go?

nowhere you don't choose. with vpc / on-device llms, requests stay inside your network. with the hosted proxy, payloads are zero-retention and never used for training.

how long does procurement take?

median 9 business days from intro to signed msa. we keep dpa and sub-processor docs current so security can self-serve.

do you support air-gapped deployment?

yes, on the enterprise+ tier. a fully on-prem agent runtime plus a model gateway for your existing llm cluster.

what your security team gets

  • ship ai coding without sending source code to a third-party model vendor.
  • give security a queryable audit.db instead of a vendor's opaque log viewer.
  • scope contractors with time-bound guest seats and per-tool permissions.
  • swap models per team without changing ides or rewriting policies.

talk to enterprise

custom volume pricing, security review packet, and a 30-day pilot for teams of 25+.