Safe AI Setup and Rescue & Harden

Safe AI Setup

Firms whose staff already paste company data into public AI tools.

About two weeks.

• An approved-tools list and a plain acceptable-use policy
• Data-classification rules: what may and may not go into AI
• A privacy impact assessment template
• Access controls and logging for the tools you choose
• Where staff use AI agents or connected tools: a list of what each may access and which actions need approval
• A one-hour session for your staff

Fixed price, agreed in writing before we start.

Leads to: The AI Readiness Audit or a care plan.

Rescue & Harden

Founders and firms with an app built mostly with AI tools that works, but is not trusted in production.

• A security and reliability review: code and dependency scanning, authentication and permission checks, secrets handling, backups, deployment and logging
• A check against the OWASP Top 10 for LLM Applications wherever the app calls an AI model
• An agent-permissions review wherever an AI agent can act
• A ranked fix list
• A fixed-price fix sprint
• A short runbook

Fixed price, agreed in writing before we start.

A review of the code and services in scope. It is not a guarantee that a system is secure.

Leads to: A care plan: maintenance and monitoring.