Machine Learning Engineer Job Description (2026): Duties, Skills and Salary
A machine learning engineer builds the systems that train, deploy and run machine learning models in production. Where a data scientist might prove a model works in a notebook, an ML engineer makes it serve real users: reliable data pipelines, training at scale, low-latency inference, drift monitoring and retraining.
The role now spans two worlds: classic ML such as ranking, recommendations, fraud and forecasting, and applied work with large language models, such as retrieval-augmented generation, fine-tuning and evaluation. ML engineers write production code in Python and often Go, Java or C++, alongside data scientists, data engineers and backend teams.
Machine Learning Engineer job description template
Copy it, replace everything in [brackets] and cut what doesn't apply. Keep the requirements to what the job really needs.
Template · replace the [brackets]
Machine Learning Engineer
[Company] · [City and office days, or remote with the countries or time zones you hire in] · [Full-time]
About [Company]
[Two or three sentences on what your company does, who it serves and why this team matters now.]
About the role
We're hiring a machine learning engineer to build and run the models behind [feature, e.g. search ranking or document understanding]. You'll take models from prototype to production, keep them accurate and fast, and build the tooling that makes the next model easier to ship.
What you'll do
- Train, evaluate and deploy models for [use case] with [PyTorch, TensorFlow or scikit-learn].
- Build feature and training pipelines together with data engineering.
- Serve models behind APIs within latency and cost targets on [cloud, Kubernetes or a managed ML platform].
- Monitor production models for drift and degradation, and automate retraining.
- Build evaluation sets, and offline and online metrics that reflect real user value.
- For LLM features: design retrieval, prompts and evaluations, and judge when fine-tuning is worth it.
- Write clean, tested production code and review other people's ML code.
What you'll bring
- [X]+ years building and deploying machine learning models in production.
- Strong Python, plus engineering practice: testing, code review, CI/CD.
- Solid ML fundamentals: model selection, overfitting, evaluation metrics.
- Experience with a deep learning framework such as PyTorch.
- MLOps tooling experience (MLflow, Kubeflow, SageMaker, Vertex AI or similar).
- Judgment in weighing model accuracy against latency, cost and complexity.
Nice to have
- LLM applications: embeddings, vector search, RAG and evaluation.
- Distributed training and GPU optimization.
- Publications or open-source contributions in machine learning.
- Recommendation or ranking systems at scale.
What we offer
- [Pay range: e.g. $X–$Y base salary, plus bonus or equity if offered]
- [GPU budget for experiments, conference attendance, learning budget]
- [Health cover, retirement or pension plan, paid time off, parental leave]
- [Work authorization: whether you can sponsor a visa or hire through an employer of record]
[Our process: a recruiter call, a coding interview, an ML system design interview, a deep dive into a model you shipped, and a team call.]
[Your equal opportunity statement, and how candidates can ask for adjustments during the process]
Put the pay range in the ad. Several US states, including California, Colorado, New York and Washington, require it in job postings, and the EU Pay Transparency Directive gives applicants the right to know the starting pay or its range before the interview.
What real machine learning engineer job ads ask for
The skills named most often in the 478 newest machine learning engineer job ads we analyzed (out of 3,613 live postings), as the share of ads that mention each. A skill most ads name is a fair must-have; one few ads name belongs under nice-to-have.
- 1Python81%
- 2LLMs69%
- 3AWS36%
- 4RAG35%
- 5CI/CD30%
- 6PyTorch28%
- 7Observability27%
- 8Azure26%
Degrees mentioned most: Computer Science (37%), Bachelor's degree (27%), Master's degree (15%).
From live job ads · computed November 2, 2026 · refreshed daily
Machine Learning Engineer duties by seniority
What the job usually looks like at each level. Pitch the ad at one level and say which in the title.
Junior
- Implements and evaluates models under guidance.
- Builds parts of the training and serving pipelines.
- Learns MLOps tooling and production practices.
Mid-level
- Owns models end to end, from data to production monitoring.
- Picks the modeling approach and evaluation method for new problems.
- Makes serving faster, cheaper and more reliable.
Senior
- Designs the ML platform and architecture across use cases.
- Sets evaluation standards and the roadmap with product leaders.
- Mentors engineers and decides where ML is, and isn't, worth it.
Machine Learning Engineer career path
- Junior ML Engineer
- Machine Learning Engineer
- Senior ML Engineer
- Staff ML Engineer or ML Engineering Manager
ML engineers usually come from software engineering with added machine learning training, or from data science with added engineering skills; graduate degrees are common but not required. Senior paths include ML platform engineering, applied research, technical leadership of AI products and engineering management.
Advertised pay for machine learning engineer jobs
Median advertised base salary in Machine Learning Engineer job ads in the United States: $222,800 (middle half: $185,000–$264,600), from 686 job ads with a published salary in the last 90 days.
This is advertised pay, not total compensation: bonuses, commission, equity and benefits aren't included, and many ads publish no pay at all.
- Machine Learning Engineer salary in the United States · median $222,800, 686 ads
- Remote Machine Learning Engineer salary · median $212,600, 336 ads
Similar roles
- Data Scientist job description
Focuses on analysis, experiments and model development more than production systems.
- Data Engineer job description
Builds the data pipelines ML systems depend on.
- Backend Developer job description
Builds the product services that call the models.
- Software Engineer job description
The general engineering role; ML engineering specializes it.
- Cloud Engineer job description
Runs the infrastructure, GPUs included, that ML workloads use.
Machine Learning Engineer job description FAQ
What does a machine learning engineer do?
Builds and runs the systems that put machine learning models into production: training pipelines, model serving, monitoring and retraining. More and more, it also means building features on large language models, with retrieval, evaluation and fine-tuning.
Machine learning engineer vs data scientist: what's the difference?
Data scientists focus on analysis, experiments and developing models; ML engineers focus on making models work reliably at scale in production. The overlap is large, so the job description should say where the balance lies.
Should an ML engineer job description ask for LLM experience?
Only if the role involves it. If you're building on large language models, ask for retrieval, evaluation and prompt or fine-tuning experience. For ranking, forecasting or fraud work, depth in classic ML matters more.