Working at Infersens
Embedded ML Engineer
InferSens Ltd is a Cambridge, UK based startup, focussed on the development of smart sensor networks for the built environment.
We are looking for an embedded ML engineer to help develop the next generation of ultra-low power sensors with on-device ML.
You will be working with leading edge silicon and will help drive design decisions relating to system partitioning, model design and implementation.
The ideal candidate will be able to demonstrate experience in one or more of the following areas:
- design and development of neural networks for time series data
- optimisation techniques for ML running on processing, memory and power constrained devices (eg pruning, quantization)
- knowledge of the application of Transformers to embedded systems
- experience of collection, storage and management of disparate sensor data for use in ML
- performing data cleaning to verify or ensure data quality
- training and retraining ML systems and models as needed
- visualising data to improve models and share insights
- using statistical analysis to improve models
- identifying differences in data distribution that could affect model performance in real-world situations
- creation of training datasets ( feature engineering, data partitioning, sampling and slicing)
- model deployment (eg model compression for OTA updates),
- prior programming experience and expertise in Python (and basic ML libraries such as numpy, scipy).
- familiarity with CI/CD methods for embedded systems including HIL testing
- experience in the verification and testing of software/firmware using Test Driven Development.
- prior experience with working with a time-series database (e.g. TimescaleDB)
Infersens has offices in Cambridge, UK but remote working options are available.
The successful candidate must, by the start of their employment, have permission to work in the UK
Application closing date: 13/06/2022
- £65-75k (full time equivalent), plus additional benefits
- Part-time role (50%), flexible working