Internship MLOps Engineer resume example

• Under supervision, linked model-registry entries to training runs and feature definitions, tested canary serving against the incumbent model, and documented rollback criteria for data drift.
• The team standardized train, validate, register, and deploy steps for 17 production models; I supported research preparation, validation, and documentation under mentor review. The team removed the longest wait from the end-to-end workflow; confirmed with my mentor that my contribution was limited to the assigned support and validation work.
• The team implemented feature freshness, drift, and serving-parity alerts for 58 features; I supported research preparation, validation, and documentation under mentor review. The team reduced repeat incidents from the same failure mode; confirmed with my mentor that my contribution was limited to the assigned support and validation work.
• Under supervision, owned model-release validation using Docker and CI/CD tests, resolved a training-serving schema mismatch and completed rollback verification before deployment.
• Under supervision, built Python monitoring using SQL feature checks and Spark ETL logs, reduced drift investigation from 60 to 25 minutes and published the model-health dashboard.

• In a mentor-reviewed simulation, packaged a Python machine learning service in Docker and added CI/CD checks for model, schema, and dependency versions; made deployment reproducible from a recorded artifact rather than a notebook state.
• As a second supervised exercise, added monitoring for prediction errors, input drift, and inference latency; tested a rollback to the previous model and verified that feature versions remained compatible after the switch.
• Under supervision, the team added canary inference and automatic rollback to 6 online models; I supported research preparation, validation, and documentation under mentor review
• Generated synthetic source records with duplicates, late arrivals and corrected values; wrote assertions for row counts and key uniqueness, and recorded the expected effect of each case on the reported metric.
• Compared the analytical output with a manually calculated reference table, traced differences to a transformation step, and retained a data dictionary and rerun instructions alongside the corrected query.
• Owned the synthetic-data validation using a manually calculated reference, resolved duplicate-key inflation and completed a notebook that reproduces the corrected totals.

Relevant coursework: Algorithms, operating systems, databases, computer networks

• Published an independent methods note using reproducible queries, explaining the data grain, excluded records and sensitivity of the result to a changed denominator.