Bootcamp | AI-powered automated trading with CRE and Confidential Workflows

​Day 1 livesteam here

​Day 2 livestream here

​Build AI-powered workflows that can reason over sensitive data, enforce policies, and execute actions—without exposing the information that should remain private.

​Across two 90-minute sessions, you’ll use Chainlink Runtime Environment (CRE) with Confidential Workflows to build real-world workflows for:

• ​Automated Liquidation Protection

• ​AI Smart Contract Audit Firewalls

​You’ll learn not just how Confidential Workflows work, but what belongs inside the confidentiality boundary, and how to design workflows that combine AI reasoning with deterministic security controls.

​By the end of the bootcamp, you’ll be able to:

• ​Understand how Confidential Workflows, TEEs, and the Vault DON work together

• ​Identify what stays protected inside an enclave — and what does not

• ​Convert a standard CRE workflow into a confidential one

• ​Fetch secrets securely inside an enclave

• ​Make confidential HTTP calls without exposing sensitive URLs, headers, request bodies, or responses to node operators

• ​Apply least-privilege capability restrictions

• ​Protect non-public thresholds, strategy parameters, and execution policies as secrets

• ​Combine LLM reasoning with deterministic guardrails, so security and capital protection never depend solely on model output

• ​Decide when a workload genuinely requires enclave-level protection

​

​Who should attend?

​This bootcamp is designed for:

• ​Developers already familiar with CRE who want to go deeper

• ​Web3 developers building trading, lending, or risk-automation products

• ​AI agent developers working with financial workflows

• ​TradFi and enterprise developers with credential, PII, or compliance constraints

• ​Security and infrastructure engineers evaluating TEE-based confidential computing for onchain systems

​Register now and learn how to build AI-powered workflows that are private where they need to be, and verifiable where they matter.

​Day 1 — Confidential Workflow Fundamentals + AI Smart Contract Audit Firewalls

​Learn how to run smart contract security audits confidentially inside a hardware enclave.

​You’ll start with the fundamentals of CRE and Confidential Workflows, including:

• ​What a TEE is and why hardware isolation matters

• ​How the Vault DON releases secrets directly into an attested enclave

• ​The confidentiality boundary — what is protected and what is not

• ​When credentials, strategy parameters, and proprietary logic should be handled inside an enclave

​Then we’ll walk through the AI Audit Firewall template end to end.

​Day 2 — Automated Liquidation Protection

​Learn how to build a confidential, AI-driven defense system for DeFi lending positions.

​You’ll first cover the mechanics behind liquidation risk, including:

• ​Loan-to-value

• ​Liquidation thresholds

• ​Health factors

• ​Why public defense thresholds can be front-run

​At the end we'll work through the Automated Liquidation Protection workflow.

​Every demo runs against a local mock server, so you can complete the bootcamp without a real LLM API key or Confidential Workflows Private Beta access.

​Before the bootcamp make sure you have:

• ​Node.js v20+

• ​Bun v1.3+

• ​CRE CLI with cre login completed

• ​Git

• ​The cre-templates repository cloned

​Optional for the onchain write exercise:

• ​Ethereum Sepolia configured in your wallet

• ​Sepolia ETH from the Chainlink faucet