Developer Joy

TPF lets teams write focused Java business functions while the framework builds, checks, and runs the repeated Quarkus pieces around them.

At a Glance

Start Fast · Design in Canvas or YAML and generate a runnable Quarkus baseline.

Type Safety · Explicit input/output types catch function, mapper, and operator mismatches at build time.

Less Boilerplate · REST, gRPC, local, and cloud-function callers are generated for you.

Use This When

• You want new developers productive quickly.
• Refactors are becoming risky because Java types and API shapes drift across modules.
• Teams are spending too much time on repeated integration glue.

In practice, developers write the domain function and the supporting code that matters: Java types, optional mappers, and clear business decisions. TPF handles the repeated Quarkus pieces around that function, including generated callers, generated runtime files, and build-time validation.

Operator Reuse

When teams already have stable Java compute libraries, operators let them plug those methods directly into the pipeline flow from pipeline.yaml. This shortens delivery time and avoids duplicate implementations.

Business Rejections Without Workflow Collapse

Not every failed item is a platform error. TPF lets step authors model per-item rejection as an expected business path using Item Reject Sink, the built-in reject-and-continue mechanism for bad records. Teams can reject specific records, continue processing the rest of the workload, and keep an audit trail plus a clear replay path that can survive process restarts when backed by a persistent provider. This avoids custom side channels and keeps recovery logic explicit in step code.

Jump to Guides

• Quick Start
• Canvas Guide
• Operators
• Item Reject Sink
• Mappers and DTOs
• Testing with Testcontainers