skills.rest
Unique-AG
Unique-AGOFFICIAL
Explore all skills by this author →
GitHub

model-deployment

Automate end-to-end LLM model deployment across environments with LiteLLM and Terraform.

5|3|Updated Jun 18, 2024
Search Tags:#llm#azure#terraform#gitops#deploy#toolkit#model-registry
One-click install
npx skills add https://github.com/Unique-AG/ai --skill model-deployment-unique-ag
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-deployment
Source: https://github.com/Unique-AG/ai/tree/main/.claude/skills/model-deployment
Command: npx skills add https://github.com/Unique-AG/ai --skill model-deployment-unique-ag

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Orchestrates the end-to-end deployment and validation of new LLM models across the Unique platform, coordinating LiteLLM (GitOps), Azure (Terraform + node backend), and the ai toolkit to prevent misconfigurations and outages.

Core Features & Use Cases

  • End-to-end rollout across environments (Dev/QA/UAT/Prod) using LiteLLM proxy config, Terraform deployments, and node-chat integration.
  • Toolkit registration and registry updates to ensure models are discoverable and usable by the language model registry.
  • Troubleshooting guidance for "model not available" errors, pricing, token limits, and provider identifiers with explicit source citations.

Quick Start

Configure LiteLLM GitOps overlays and Azure deployments for a new model, register it in the toolkit, and run end-to-end validation across environments.

Frequently Asked Questions about model-deployment

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

FAQPage Schema
How do I automate LLM deployment across multiple environments using Terraform and GitOps?▼

Automating LLM deployment involves orchestrating LiteLLM proxy config via GitOps and Azure resources via Terraform to roll out models sequentially across Dev, QA, UAT, and Prod environments while preventing misconfigurations.

What is the best way to register a new LLM in a model registry and make it discoverable?▼

Registering a new LLM requires updating the ai toolkit and language model registry during the deployment process, ensuring the model is discoverable and usable by collecting pricing, token-limit data, and provider identifiers from authoritative sources.

Why does my deployed LLM show a model not available error after rollout?▼

A model not available error after rollout often indicates a misconfiguration in the LiteLLM GitOps overlays, incomplete Terraform Azure deployments, or a missing node-chat integration that requires troubleshooting using cited provider documentation.

Does this deployment approach support collecting token limits and pricing data from official docs?▼

Yes, the deployment process explicitly collects pricing, token-limit data, and provider identifiers from authoritative sources, requiring citations from official documentation and Jira or PR references to validate the model configuration.

How does end-to-end LLM validation work across Dev, QA, UAT, and Prod environments?▼

End-to-end validation works by coordinating LiteLLM proxy configuration, Terraform infrastructure deployments, and node-chat integration to ensure the LLM passes rollout checks across Dev, QA, UAT, and Prod without outages.

Related Skills

Explore More Software Engineering Skills→
jonathan-vellajonathan-vella
6

azure-validate

Validate Azure environments and configurations before deployment using Bicep or Terraform.

Community
Advanced
Pupfish-LLCPupfish-LLC
1

qlik-deploy

Automate Qlik Sense app deployment across Cloud and client-managed environments.

Official
Advanced
jackspacejackspace
16

ai-elements-chatbot

Compose streaming AI chat interfaces with Vercel AI SDK v5 components.

Community
Advanced
laurigateslaurigates
10

Infrastructure Terraform

Automate cloud infrastructure provisioning with Terraform and HCL.

Community
Advanced