Dear Hiring Manager,
After several years in an adjacent role, I am moving deliberately into ml engineer work because the strongest part of my recent job has been training-serving consistency, latency work, drift monitoring, and release controls. Over the last year I have built hands-on evidence with Python, PyTorch, Spark, Airflow, feature stores, Docker, Kubernetes, and MLflow, so this is a planned move rather than a loose interest.
I am not asking you to infer the connection between my old title and this role. Your team needs production ML judgement, model serving, pipeline reliability, and monitoring discipline, and my strongest examples sit in that exact area. I would use this letter to show the connection with one specific project, the constraints I worked under, and the judgement I brought to the decision points.
A recent example is that I moved a fraud model from notebook handoff to monitored serving with feature freshness checks and shadow evaluation. That work required Python, PyTorch, Spark, Airflow, feature stores, Docker, Kubernetes, and MLflow, but the more important point is how I made decisions, explained tradeoffs, and followed the result through after release.
I would welcome the chance to talk through how this transition maps to your team needs. ML engineering teams need evidence that models survive production traffic, so I would keep the letter concise, evidence-led, and tied to the outcomes the hiring team is likely to care about.
Yours sincerely, Alex Morgan