Use Cases

Tangle supports AI-native products and infrastructure where work needs isolation, auditability, and reliable payments. These examples show how the workbench, sandbox runtime, and protocol fit together.

Agentic Software Engineering

Run background agents that build features, refactor code, write tests, and open pull requests. Work stays isolated and reviewable. Good for: product teams using the workbench with strict review gates. Layers: workbench + runtime.

Evaluation and Governance Pipelines

Execute large evaluation suites across prompts, models, and tools. Each run produces structured metrics so teams can track regressions and enforce policies. Good for: AI builders who need repeatable evaluations and policy enforcement. Layers: workbench + runtime.

Regulated and Sensitive Workflows

Deploy AI workflows on protected data with strict isolation, resource limits, and audit logs. Outputs are verifiable and reviewable. Good for: regulated teams that require strong isolation and auditability. Layers: runtime + protocol (for paid operators).

Data and Knowledge Operations

Use Blueprints to run extraction, transformation, labeling, and retrieval jobs. Operators provide compute while the protocol coordinates payments and accountability. Good for: teams turning knowledge work into reliable, reusable services. Layers: Blueprint SDK + protocol.

AI Ops and Reliability

Operate autonomous monitoring, cost optimization, and incident response workflows. Operators are paid for uptime and performance, while services can enforce reliability targets. Good for: operators and infrastructure engineers running long-lived services. Layers: runtime + protocol.

Marketplace-Ready AI Services

Package an agent workflow or infrastructure service as a Blueprint, publish it once, and let others instantiate it on demand. Payments flow to operators and developers based on usage. Good for: builders who want distribution without running their own infrastructure. Layers: Blueprint SDK + protocol.