An orchestration role that coordinates AI/ML model work through associated skills rather than performing hands-on tasks.

Does AIEng perform tasks directly?

No; tasks are carried out by its linked skills (MLOps, LLM integration, evaluation, etc.).

How do I work with AIEng?

Define the project scope and ensure each lifecycle stage is mapped to the appropriate skill; DataLead approves and reviews results.

Aieng

Scanned

npx machina-cli add skill javalenciacai/develop-skills/aieng --openclaw

Files (1)

SKILL.md

2.7 KB

AIEng - AI Engineer

Role

Develops and implements AI/ML models. Reports to DataLead.

Responsibilities

AI and machine learning model development and implementation
Model training and fine-tuning
MLOps and model deployment in production
LLM and generative model integration
Model evaluation and monitoring
Critical Restriction: This skill is only a role and must always use one of its associated skills. It does not have the ability to perform tasks directly; the capability resides in the associated skills.

Base Skills

# Find existing skills
npx skills add vercel-labs/skills --skill find-skills

# Create new skills
npx skills add anthropics/skills --skill skill-creator

Current Skills

Base Skills (All AI Engineers)

Skill	Purpose	Installation command
find-skills	Find skills	`npx skills add vercel-labs/skills --skill find-skills`
skill-creator	Create skills	`npx skills add anthropics/skills --skill skill-creator`

AI/ML Development Skills 🔴 High Priority

Skill	Purpose	Installation command
doc-coauthoring	Model documentation, ML specs, training procedures, model cards, evaluation reports	`npx skills add anthropics/skills --skill doc-coauthoring`
mcp-builder	LLM integration, MCP servers, AI service architecture, API design for ML models	`npx skills add anthropics/skills --skill mcp-builder`
xlsx	Model performance tracking, experiment logs, hyperparameter tuning, ML metrics	`npx skills add anthropics/skills --skill xlsx`

Documentation and Knowledge Sharing Skills 🟡 Medium Priority

Skill	Purpose	Installation command
technical-blog-writing	ML best practices, model insights, AI tutorials, MLOps guides	`npx skills add 1nference-sh/skills --skill technical-blog-writing`

Rule: Add Used Skills

Every time you use a new skill, add it to the "Current Skills" table.

Examples of skills to search for:

npx skills find machine-learning
npx skills find llm
npx skills find mlops

Source

git clone https://github.com/javalenciacai/develop-skills/blob/main/.agents/skills/aieng/SKILL.mdView on GitHub

Overview

AIEng develops and implements AI/ML models across training, deployment, and monitoring. It handles LLM integration, prompt engineering, and generative AI features, including RAG systems. This role orchestrates work via associated skills rather than performing tasks directly.

How This Skill Works

AIEng does not execute tasks by itself; it orchestrates the work through specialized skills (e.g., mcp-builder for architecture, xlsx for metrics, doc-coauthoring for docs, and LLM integration tools). It defines model objectives, training regimes, evaluation criteria, and deployment plans, then delegates execution to the appropriate skill and aggregates results for DataLead.

When to Use It

Developing or training AI/ML models
MLOps, model deployment or model monitoring
LLM integration or prompt engineering
Generative AI features or RAG systems
Model evaluation, hyperparameter tuning or A/B testing

Quick Start

Step 1: Define objective, data sources, and success metrics
Step 2: Select associated skills (e.g., mcp-builder for deployment, doc-coauthoring for docs, xlsx for metrics) and design the workflow
Step 3: Execute via the chosen skills and review outcomes with DataLead

Best Practices

Define clear model objectives and success metrics before starting
Map each lifecycle stage to the appropriate associated skill
Enforce versioning, reproducibility, and strict access controls
Implement continuous monitoring, drift detection, and alerting
Document model cards, data provenance, and architectural decisions

Example Use Cases

Orchestrate a sentiment-analysis model from data prep to deployment using mcp-builder and monitoring
Integrate a chat-based LLM into a customer support workflow with prompt engineering and MCP servers
Conduct hyperparameter tuning and A/B testing with defined evaluation criteria
Build a RAG-powered retrieval feature for a knowledge base and monitor its performance
Evaluate a generative image model and rotate deployment with MLOps pipelines

Frequently Asked Questions

Add this skill to your agents