Machine Learning Engineer

Machine Learning Engineer WHAT WE DO Gecko Robotics is helping the world’s most important organizations ensure the availability, reliability, and sustainability of critical infrastructure. Gecko's complete and connected solutions combine wall-climbing robots, industry-leading sensors, and an AI-powered data platform to provide customers with a unique window into the current and future health of their physical assets. This enables real-time decision making to increase the efficiency and safety of operations, promote mission readiness, and protect the environment and civilization from the effects of infrastructure failure. ROLE AT A GLANCE As a Machine Learning Engineer at Gecko, you will be working with Gecko’s unique dataset to develop and deploy machine learning models to solve critical business problems. You will work deeply on problems such as: classifying valid vs. invalid signals; taking accurate measurements from valid signals; and identifying damage mechanisms such as cracks, corrosion, or laminations across a large collection of signals. Gecko owns a growing repository of mechanical integrity data, including large volumes of ultrasonic, imagery, and other data points concerning the integrity of critical infrastructure assets worldwide. Gecko is expanding our Machine Learning team to better leverage the vast data store we have collected over time. This is a chance to be one of the first dedicated engineers in an area we expect to grow significantly over the next few years, within a space that is ripe for innovation and solving problems in a way they haven’t been attempted before. You will be a key member of our advanced technology team, responsible for pioneering novel machine learning models that interpret complex sensor data from our robotic inspection platforms. This hybrid role bridges the gap between state-of-the-art research and practical engineering. WHAT YOU WILL DO - Design, develop, and deploy sophisticated machine learning models (supervised and unsupervised) to analyze complex sensor data for non-destructive evaluation (NDE). - Lead research and implementation of next-generation, physics-based machine learning models that can interface with or replace traditional simulation methods like Finite Element Analysis (FEA) or Computational Fluid Dynamics (CFD). - Process and interpret large-scale, unconventional datasets, including raw ultrasonic A-scans and infrared thermal images, to extract meaningful insights. - Collaborate with robotics, software, and data engineering teams to integrate ML solutions and algorithms into our production data platform. - Stay at the forefront of academic and industry advancements in machine learning, signal processing, and physics-based modeling to drive innovation. - Own the end-to-end lifecycle of ML projects, from problem formulation and data exploration and curation to model deployment, monitoring, and iteration. Technologies We Use - Python, PyTorch, Numpy, Scipy, Pandas, Scikit-learn - Docker, K8s, GCP, Cloud Run, Batch, Collab - Ultrasonic DSP We use a variety of technologies, but our Software teams primarily operate using Python, React, and Typescript with Google Cloud Platform (GCP) as our cloud provider. This is a non-exhaustive list and we are tech agnostic in our interview process, so we encourage you to apply regardless of your background. ABOUT YOU Required Skills - Bachelor’s degree in Computer Science or a closely related field, or equivalent practical experience. - 5+ years of professional engineering experience, with at least 3+ years in a dedicated machine learning role. - Strong practical knowledge of machine learning algorithms and the ability to read and implement relevant research papers, especially for time-series analysis and anomaly detection in signal data. - Proficiency in Python and at least one major machine learning framework (e.g., PyTorch, TensorFlow), with experience deploying at least one model into a production environmen ... (truncated, view full listing at source)