Pi-based coding agent with hard budget caps

Kimchi launched a governed AI coding platform that enforces hard budget caps, real-time spend attribution, and zero model-data egress before any request reaches a model, installable via a curl script from its GitHub releases page. The platform lets organizations set spending limits by user, team, API key, or model, deploy in Kubernetes on AWS, GCP, Azure, or on-premises, and excludes prompts, code, and outputs from model training. Kimchi said the CLI installs with no credit card and no lock-in, with spending policy enforced before a request leaves the customer's environment.

Hard budget caps, real-time spend attribution, and zero model-data egress - enforced before a request hits a model. curl -fsSL https://github.com/getkimchi/kimchi/releases/latest/download/install.sh | bash Every agent request passes through one enforcement layer for spending, data, deployment, and model routing. Set caps by user, team, API key, or model. Spending stops at the limit. Deploy in Kubernetes on AWS, GCP, Azure, or on-premises. Keep model data inside. Prompts, code, and outputs are excluded from model training. Set spending limits, track every dollar, and keep your code and data within your boundaries - all in one platform. See exactly who, what, and where your AI spend is going. Per-developer, per-team, per-model, per-tag - in real time. Hard budget caps at every level: user, team, API key. One governed AI coding platform, built around the way each team already works. Platform engineering Start serverless or deploy inside your VPC. Developers Use Kimchi from the coding tools your team already relies on. Security teams Spending policy enforced before a request leaves your environment. Security, compliance, and control aren't afterthoughts. Every feature ships enterprise-ready. Install the CLI, run /ferment, and hand off to the agent. No lock-in, no credit card.