ml-engineer

Design and deploy ML models with reproducible training pipelines and monitoring.

Updated Apr 4, 2026

Search Tags:#deployment #mlops #drift-detection #inference #model-training #ml-engineering

Ready to run

npx skills add https://github.com/asalhamed/dev-agents --skill ml-engineer-asalhamed

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ml-engineer
Source: https://github.com/asalhamed/dev-agents/tree/main/ml-engineer
Command: npx skills add https://github.com/asalhamed/dev-agents --skill ml-engineer-asalhamed

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Design and deploy robust ML models by bridging research and production, ensuring reproducibility and scalable deployment.

Core Features & Use Cases

Model design & architecture: select architectures for classification, detection, anomaly detection, or time-series tasks with explicit input/output contracts.
Training pipelines & experiments: reproducible pipelines with versioned data, seeds, and experiment tracking; supports MLflow-style logging and model registry.
Deployment & monitoring: export formats (ONNX, TorchServe), monitoring for drift and performance, edge and cloud targets.

Quick Start

Set up a reproducible ML project that defines problem, data schema, training plan, deployment target, and monitoring strategy.

Frequently Asked Questions about ml-engineer

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

FAQPage Schema

How do I build reproducible ML training pipelines for model deployment?▼

Reproducible ML training pipelines require versioned data, fixed seeds, and experiment tracking to ensure consistent results before model deployment. This workflow supports MLflow-style logging and model registry integration to bridge research and production.

What is the best way to export ML models for ONNX and TorchServe deployment?▼

Exporting ML models for ONNX and TorchServe deployment involves converting trained artifacts into standard formats for scalable inference. This ensures models are portable across cloud and edge targets while maintaining explicit input and output schemas.

How does model monitoring and drift detection work in MLOps pipelines?▼

Model monitoring and drift detection in MLOps pipelines track performance degradation and data shifts post-deployment. This process evaluates incoming inference data against baseline distributions to trigger alerts when model behavior diverges from training expectations.

Can I use this approach for both cloud and edge ML deployments?▼

Yes, this approach supports both cloud and edge ML deployments by defining explicit deployment targets early in the project lifecycle. It accommodates diverse architectures including predictive modeling, anomaly detection, and computer vision tasks.

What input schemas and privacy considerations are needed for ML engineering tasks?▼

ML engineering tasks require explicit input and output contracts, defining data schemas and privacy considerations before training begins. This structured framing ensures reproducible pipelines and compliant model deployment across predictive modeling and monitoring workflows.