ML Intern: Optimization Quality & Monitoring

Nexoya Ltd.; Zürich; On-site; CHF 42,000 - 54,000; Full time

Job summary

Nexoya Ltd. in Zürich is seeking a motivated ML Intern to join our team building AI-driven budget optimisations and simulations for enterprise marketing portfolios. You will monitor prediction pipelines, validate results, and turn recurring checks into reliable tooling.

Work with ML engineers and Customer Success to understand how your work shapes customers' budget decisions. Fluency in English is required; German or Italian is a bonus. Basic Python, SQL and statistics knowledge is expected.

Qualifications

• Master's degree in Data Science, CS, Stats, Math, Engineering or related field.
• Strong Python skills with pandas/NumPy.
• Fluent English, spoken and written.
• Ability to explain technical findings to non-technical colleagues.

Responsibilities

• Monitor automated prediction, optimisation and simulation pipelines across portfolios and track run health, completeness and timeliness.
• Verify that predictions, optimisation proposals and simulation results meet our quality standards before they reach customers.
• Investigate failed or unexpected runs, identify the root cause and resolve or elevate with a clear diagnosis.
• Define and maintain plausibility checks for optimisation and simulation results.
• Build dashboards showing run status, data health and model performance across portfolios.

Skills

Python skills

Statistics knowledge

Machine learning concepts

English fluency

Attention to detail

Education

Master's degree in a related field

Tools

Python

SQL

pandas/NumPy

Git

Docker

Job description

Our common vision

Our mission is to help marketing teams unlock their full growth potential with AI. Our AI-powered attribution creates real transparency, showing how digital advertising investments perform across channels. On that foundation, we optimise budgets so our customers can sustainably grow their bottom line. We support leading European brands across e-commerce, telco, banking, insurance & energy, and many other industries in making their marketing smarter, more efficient, and more impactful.

About the role

Nexoya runs fully automated, AI-driven budget optimisations and simulations for enterprise marketing portfolios across many ad channels. Every result that reaches a customer must be correct, explainable and on time. As our ML Intern, you make sure it is: you monitor pipeline health, check the quality of predictions, optimisations and simulations, investigate anomalies and turn recurring manual checks into reliable tooling. You work closely with our ML engineers and Customer Success, and you see directly how your work shapes our customers' budget decisions.

Your responsibilities

Monitor the automation and assure result quality

• Monitor the automated prediction, optimisation and simulation pipelines across customer portfolios and track run health, completeness and timeliness.

• Verify that predictions, optimisation proposals and simulation results meet our quality standards before they reach customers: budget shifts, expected KPI changes, channel constraints and funnel-level effects.

• Investigate failed or unexpected runs, identify the root cause (data, configuration or model) and resolve it or elevate it with a clear diagnosis.

Assure the quality of results over time

• Define and maintain plausibility checks for optimisation and simulation results, for example for budget planning, seasonal shifts or new channel tests.

• Make sure results are consistent and explainable, so that customers and account managers can trust them without manual checks.

• Compare optimised and simulated outcomes with realised performance, document deviations and feed the findings back to the ML team.

Safeguard data and model quality

• Monitor input data for gaps, anomalies and tracking changes that affect optimisation quality.

• Track model fit and prediction accuracy per portfolio and flag degradation early.

• Maintain portfolio configurations such as funnel steps, constraints and target definitions together with the responsible teams.

Automate the operational workflow

• Replace manual checks with Python scripts, validation rules and alerts.

• Build and maintain dashboards that show run status, data health and model performance across portfolios.

• Write runbooks so that operational knowledge is shared, not held by one person.

The impact you will have

• Customers receive optimisation proposals and simulations that are correct and delivered on time.

• Issues are caught before customers see them, not after.

• The time the ML team spends on operational work goes down measurably, because recurring steps are automated.

• Customer Success can trust automated optimisation and simulation results without checking them by hand.

Your experience and skills

• Recently completed Master's degree in Data Science, Computer Science, Statistics, Mathematics, Engineering or a related field.

• Solid Python skills, including pandas and NumPy, and comfort working with SQL.

• Good understanding of statistics and basic machine learning concepts such as regression, model evaluation and overfitting.

• A careful, structured way of working: you check results before you share them and you document what you do.

• Ability to explain technical findings clearly to non-technical colleagues.

• Fluent English, spoken and written, is a must.

Bonus points

• Experience with optimisation methods, for example constrained or non-convex optimisation.

• Familiarity with Marketing Mix Modelling, attribution or digital advertising metrics.

• Experience with data pipelines, scheduled jobs or monitoring tools.

• Experience with Git, Docker and cloud environments.

• Fluent in German or Italian.

Your future work life

• Purpose: Your work directly influences how enterprise marketers allocate their budgets.

• Responsibility: You own the quality of a defined part of our automation from day one, with a personal development roadmap and regular feedback.

• Learning: You work alongside experienced ML and data engineers on production systems, not on side projects.

• Culture: A transparent company culture at eye level, where every person is valued and supported.

• Strategy: You take part in shaping company strategy and culture during our quarterly Strategy Days.

• Events: Regular team events, on-site and off-site.

Application process

• Profile review and short phone screen

• Take-home case study

• Case discussion with the ML team

• On-site interview and team evaluation