Which orchestration platforms are supported?

The skill currently supports Apache Airflow, Prefect, and standard Cron scripts to accommodate different infrastructure requirements.

Can I use it with TimeGPT?

Absolutely. It is designed to work seamlessly with Nixtla's TimeGPT as well as open-source libraries like StatsForecast.

What prerequisites are needed before generating a pipeline?

You should have a completed experiment configuration file (usually created via nixtla-experiment-architect) and the necessary environment variables like NIXTLA_API_KEY set.

Does it include model monitoring?

Yes, it generates monitoring scripts that track accuracy metrics like sMAPE and MASE, with built-in logic to trigger alerts or fallbacks.

How does the fallback logic work?

If the primary forecast quality check fails or the API is unavailable, the pipeline is configured to automatically revert to baseline statistical models to ensure service continuity.

Nixtla Production Pipeline Generator

Name: Nixtla Production Pipeline Generator
Author: intent-solutions-io

byintent-solutions-io

0•

Despliegue y DevOps

Converts Nixtla forecasting experiments into production-ready Airflow or Prefect pipelines with integrated monitoring and error handling.

The Nixtla Production Pipeline Generator bridges the gap between experimental data science and production engineering by automating the creation of robust inference pipelines. It transforms validated forecasting configurations into complete deployment artifacts, including Airflow DAGs, Prefect Flows, or simple Cron scripts. Designed for enterprise reliability, the skill implements a standardized Extract-Transform-Forecast-Load (ETFL) pattern and includes automated performance monitoring with sMAPE/MASE metrics and fallback logic to ensure forecasting continuity even when primary models fail.

Características Principales

01Automated Airflow DAG and Prefect Flow generation

02Integrated performance monitoring and quality checks

030 GitHub stars

04Standardized ETFL pipeline architecture

05Config-driven deployment from validated experiments

06Automatic fallback to baseline models on failure

Casos de Uso

01Creating lightweight, scheduled cron-based forecasting tasks for smaller projects

02Implementing automated model performance monitoring and alerting in production

03Deploying TimeGPT forecasting models to enterprise Airflow environments

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add intent-solutions-io/plugins-nixtla nixtla-prod-pipeline-generator

For use in Claude.ai and ChatGPT

Download Skill