ML Engineer是ML & AI Engineering族群中的角色。涵盖5个级别的58项技能（从Junior到Principal）。其中125项为必备技能。关键领域：Programming Fundamentals, Backend Development, Database Management。
技术栈
各级别重点
Training models using existing pipelines. Feature engineering. Model validation. Preparing datasets. Working with Jupyter notebooks.
Designing ML pipelines. Model selection and tuning. A/B testing models. Deploying models to production. Feature store.
ML systems architecture. Inference optimization (ONNX, TensorRT). Designing real-time ML. Researching new approaches. Mentoring.
ML platform strategy. MLOps infrastructure. Coordinating ML and backend. Experimentation standards. ML team roadmap.
Company AI strategy. LLM integration. ML at scale. Research agenda. Publications and talks.
技能矩阵
58 技能 × 5 级别. 点击单元格查看详情。
AI-Assisted Development
4 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| GitHub Copilot | A | A | W | A | A |
| Cursor IDE | A | W | A | A | — |
| ChatGPT / Claude | A | A | W | A | A |
| Prompt Engineering for Code | A | A | W | A | A |
API & Integration
5 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| REST API Design | A | W | A | A | E |
| GraphQL Design | A | W | A | E | E |
| WebSocket API Design | A | W | A | E | E |
| gRPC & Protocol Buffers | A | A | W | A | A |
| API Documentation | A | W | A | E | E |
Architecture & System Design
1 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| System Design Fundamentals | A | W | A | E | E |
Backend Development
4 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Python Web Frameworks | A | W | A | A | E |
| Apache Kafka | A | A | W | A | A |
| Redis | A | A | W | A | A |
| Task Queues | A | A | W | A | A |
Cloud & Infrastructure
5 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Docker | A | W | A | A | E |
| Kubernetes Core | A | W | A | A | E |
| Terraform | A | W | A | E | E |
| AWS | A | W | A | A | E |
| Network Fundamentals | A | W | A | E | E |
Data Engineering
4 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Apache Spark | A | W | A | E | E |
| Pandas / Polars | A | W | A | E | E |
| SQL-based ETL | A | W | A | E | E |
| Data Quality | A | W | A | E | E |
Database Management
4 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| PostgreSQL | A | W | A | A | E |
| Database Indexing | A | A | W | A | A |
| Query Optimization | A | A | W | A | A |
| Data Modeling & Schema Design | A | W | A | E | E |
DevOps & CI/CD
1 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| GitHub Actions / GitLab CI | A | W | A | A | E |
Machine Learning & AI
9 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Classical ML (scikit-learn) | A | W | A | E | E |
| Gradient Boosting | A | W | A | E | E |
| PyTorch | A | W | A | E | E |
| MLflow | A | W | A | E | E |
| Feature Stores | A | W | A | E | E |
| Model Serving | A | W | A | E | E |
| Experiment Tracking | A | W | A | E | E |
| ML Pipelines | A | W | A | E | E |
| Model Monitoring | A | W | A | E | E |
Observability & Monitoring
4 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Structured Logging | W | A | E | — | — |
| Prometheus & Grafana | A | W | A | — | — |
| OpenTelemetry | A | W | A | E | E |
| SLI / SLO / SLA | A | W | A | E | E |
Programming Fundamentals
9 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Algorithms & Complexity | W | W | A | A | E |
| Data Structures | W | W | A | A | E |
| OOP & SOLID Principles | W | W | A | A | E |
| Design Patterns | W | W | A | A | E |
| Multithreading | W | W | A | A | E |
| Async Programming | W | W | A | A | E |
| Code Quality & Refactoring | W | W | A | A | E |
| Type Safety & Type Systems | A | W | A | E | E |
| Memory Management | A | W | A | E | E |
Security
3 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| OWASP & Application Security | A | W | A | E | E |
| Secure Coding Practices | A | W | A | E | E |
| JWT / OAuth2 / OIDC | A | W | A | E | E |
Testing & QA
3 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Unit Testing | A | W | A | A | E |
| Integration Testing | A | W | A | A | E |
| E2E Testing | A | W | A | E | E |
Version Control & Collaboration
2 技能| 技能 | Jun | Mid | Sen | Lead | Princ |
|---|---|---|---|---|---|
| Git Advanced | A | W | A | A | E |
| Code Review | A | W | A | A | E |
常见问题
ML Engineer角色需要哪些技能？
ML Engineer角色需要58项技能，其中125项为必备。技能分布在5个级别：从Junior到Principal。 查看完整矩阵.
如何在ML Engineer角色中晋升到下一级别？
使用等级计算器评估您当前的级别并获取个性化建议。系统将显示晋升所需发展的技能。
ML Engineer角色使用什么技术栈？
技术栈包含5种不同级别的技术。 Python 3.11+, scikit-learn, pandas/numpy, Jupyter, PyTorch/TensorFlow basics, SQL, Git, Python 3.12+, PyTorch/TensorFlow, XGBoost/LightGBM/CatBoost, MLflow, Airflow, Feature store basics, Docker, DVC, PyTorch advanced, ONNX/TensorRT, Triton Inference Server, Kubeflow/MLflow, Spark ML, Ray, Kubernetes, Prometheus/Grafana...
社区如何定义ML Engineer角色的要求？
角色要求由社区通过提案系统制定。任何成员都可以提出修改建议，经过投票和专家评审后生效。