ML/Data Infrastructure Engineer

Flyability, a Swiss scale‑up that builds collision‑tolerant inspection drones, is looking for an ML/Data Infrastructure Engineer to join its Autonomy team in Lausanne. The company's mission is to send robots into hazardous, confined spaces instead of people, and its flagship product Elios already serves thousands of customers worldwide. In this role you will design, implement and run the data and machine‑learning pipelines that turn raw drone footage into usable AI models. You will be responsible for the end‑to‑end workflow: ingesting raw sensor streams, organising and versioning datasets, hosting internal labeling tools, automating training and evaluation environments, tracking experiments, and delivering models to production. Close collaboration with Spatial AI engineers and the Cloud Platform Engineer is essential to keep the whole stack reliable and reproducible. The position requires at least three years of professional experience in data engineering, MLOps or a closely related software‑engineering field. You must be proficient in Python and comfortable building production‑grade systems. Hands‑on expertise with data storage solutions, databases and large‑scale data processing is mandatory, as is practical knowledge of AWS services such as S3 and cloud compute resources. You should have worked with the full ML lifecycle, dataset and model versioning, experiment tracking, training orchestration, model registries or CI/CD for ML. Experience handling ML datasets, including curation, annotation and quality assurance, as well as familiarity with Docker, CI/CD pipelines, workflow orchestration tools and infrastructure‑as‑code, is also required. Strong English communication skills are a must; French is a plus. Additional, non‑essential skills that would be valuable include familiarity with MLOps platforms like MLflow, DVC or SageMaker, experience running annotation platforms and coordinating external labeling teams, and the ability to deploy models on embedded or resource‑constrained hardware. Experience building pipelines for fine‑tuning, evaluation and low‑latency serving of large language models is also welcomed. The role offers a flexible schedule with up to two remote days per week, 25 vacation days plus public holidays, accident insurance, and various wellness and community benefits. What the role asks for: - 3+ years data engineering / MLOps experience - Strong Python programming skills - Hands‑on data storage, database, large‑dataset processing - Practical AWS experience, especially S3 and cloud compute - Experience with ML lifecycle (versioning, experiment tracking, CI/CD) - Experience with ML dataset curation, annotation, quality - Proficiency with Docker, CI/CD, workflow orchestration, IaC - Proficiency in English; French a plus - Nice-to-have: MLflow, DVC, SageMaker experience - Nice-to-have: annotation platforms and external labeling teams - Nice-to-have: deploying models to embedded/resource‑constrained devices - Nice-to-have: pipelines for fine‑tuning and low‑latency LLM serving