Científicos de datos

Ocupaciones informáticas y matemáticas · Habilidades y competencias laborales · SOC 15-2051 · 44.994 skills

Contexto de la ocupación

Estas tareas de O*NET describen la ocupación y aportan contexto a la asignación SOC; no significan que cada Skill de la lista realice todas las tareas.

Analyze, manipulate, or process large sets of data using statistical software.
Apply feature selection algorithms to models predicting outcomes of interest, such as sales, attrition, and healthcare use.
Apply sampling techniques to determine groups to be surveyed or use complete enumeration methods.

Las tareas de O*NET se muestran en su idioma de origen (inglés). Los datos de tareas se han extraído de O*NET® 29.1 Database de USDOL/ETA y se utilizan bajo CC BY 4.0.

Fuentes representativas

Una muestra limitada de Skills populares asignadas a esta ocupación, con un máximo de una por repositorio de GitHub. A continuación aparece la lista completa.

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

jupyter-notebook

NousResearch/hermes-agent

Iterative Python via live Jupyter kernel (hamelnb).

Todas las Skills asignadas

Explora la asignación completa por popularidad del repositorio o actividad reciente de la fuente.

Los nombres y las descripciones conservan el idioma proporcionado por el repositorio de origen.

Ordenar por

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

pytorch-patterns

日本語翻訳：このファイルは pytorch-patterns 用の日本語翻訳が必要です

pytorch-patterns

PyTorch深度学习模式与最佳实践，用于构建稳健、高效且可复现的训练流程、模型架构和数据加载。

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

pytorch-patterns

PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading.

ml-adoption-playbook

End-to-end methodology for AI agents and software engineers to add machine learning algorithms to existing non-ML codebases. Covers problem framing, data readiness, architectural decoupling, and baseline model integration. Use when adding a machine learning capability to a codebase that has none, from problem framing through a baseline model.

Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.

pytorch-patterns

PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading. Use when writing or reviewing PyTorch training loops, model architectures, or data loading, or when a run will not reproduce.

jupyter-notebook

NousResearch/hermes-agent

Iterative Python via live Jupyter kernel (hamelnb).

evaluating-llms-harness

NousResearch/hermes-agent

lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).

weights-and-biases

NousResearch/hermes-agent

W&B: log ML experiments, sweeps, model registry, dashboards.

NousResearch/hermes-agent

Train sparse autoencoders to interpret model features.