About the role
As a Machine Learning Engineer, you will design, build, and deploy the deep learning models at the heart of Falkonry's time series AI platform. This includes anomaly detection models, model optimization for low-latency inference, and end-to-end ownership of the ML lifecycle across edge and cloud environments.
Responsibilities:
- Design, train, and rigorously test Artificial Neural Network (ANN) models to solve complex business and technical challenges.
- Research, conceptualize, and implement new methods or discover novel applications for existing methods to significantly improve our anomaly detection capabilities.
- Fine-tune, quantize, and optimize models for high-speed, low-latency inference across diverse deployment platforms and edge devices.
- Take ownership of the machine learning lifecycle, from data preparation and architecture selection to deployment and monitoring.
- Collaborate closely with the product and engineering teams to deliver end-to-end capabilities.
- Participate in code reviews, system design discussions, and incident response.
Requirements:
- Strong fundamentals in machine learning model architecture, training dynamics, and evaluation.
- Hands-on experience with deep learning frameworks (PyTorch, Keras, or similar), including custom training loops and tensor operations.
- Familiarity with core architectures such as Transformers, Variational Autoencoders (VAEs), and Convolutional Neural Networks (CNNs).
- Ability to learn quickly, work independently, and take ownership of deliverables.
- Contributions to open source or personal projects demonstrating engineering depth.
Nice to Have:
- Experience with model export and optimization for cross-platform inference (ONNX Runtime or similar).
- Familiarity with anomaly detection, time-series modeling, or generative models.
- Experience deploying ML models to edge devices or air-gapped environments.