haipipe-end

Package trained ModelInstance_Set into Endpoint_Set and deploy to Databricks or local.

1|Updated Feb 16, 2026

Search Tags:#workflow #deployment #endpoint #builder #inference #mlflow #haipipe-end

Ready to run

npx skills add https://github.com/jluo41/Tools --skill haipipe-end

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: haipipe-end
Source: https://github.com/jluo41/Tools/tree/main/plugins/research/skills/haipipe-end
Command: npx skills add https://github.com/jluo41/Tools --skill haipipe-end

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Unified workflow to package a trained ModelInstance_Set into a production-ready Endpoint_Set, enabling seamless testing, design of five endpoint-inference function types (MetaFn, TrigFn, PostFn, Src2InputFn, Input2SrcFn), and deployment to Databricks or local environments.

Core Features & Use Cases

End-to-end lifecycle: packaging, testing, Fn design, and deployment across Databricks or local platforms.
Fn-type orchestration: supports all five Fn types with a guided builder workflow and YAML config.
Deployment orchestration: handles packaging, warmup, inference, and deployment steps in a repeatable process.

Quick Start

Run the haipipe-end workflow to package a trained endpoint, validate it with tests, and deploy to your chosen platform.

Frequently Asked Questions about haipipe-end

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deploy a trained ML model as an inference endpoint?▼

You can design an ML inference endpoint by orchestrating five function types (MetaFn, TrigFn, PostFn, Src2InputFn, Input2SrcFn) alongside YAML configurations to structure the model's data processing and deployment logic.

Can I deploy my machine learning workflow to both Databricks and local environments?▼

Yes, the endpoint deployment workflow handles packaging, warmup, inference, and deployment steps to seamlessly deploy your machine learning models to either Databricks or local environments.

What inference functions do I need to configure for ML endpoint deployment?▼

You must configure five inference function types: MetaFn, TrigFn, PostFn, Src2InputFn, and Input2SrcFn. They are coordinated through a clear dispatch table and YAML configurations to structure endpoint behavior.

How do I package a trained model for production endpoint deployment?▼

You package a trained model for production endpoint deployment by converting the ModelInstance_Set into an Endpoint_Set, validating it through warmup and inference tests, and executing deployment steps via a repeatable orchestration process.

Does endpoint deployment require YAML configurations for inference functions?▼

Yes, YAML configurations are enforced during endpoint deployment to define and coordinate the behavior of the five inference function types, ensuring a repeatable packaging and testing workflow.