Summary Sheet: I.T. & Communications
| || || || |
|Advertiser Name||SafetySpect UK Limited||Advertiser Type:||Company|
|Classification:||I.T. & Communications||Subclassification:|
|Country:||United Kingdom||Location:||United Kingdom|
|Language:||English - United Kingdom (en-GB) ||Contact Name:||>SafetySpect UK Limited|
|Employment Type:||Permanent||Workhours:||Full Time|
Position: Machine Learning Specialist
SafetySpect UK limited is developing optical sensing technology that help identify species of fish and freshness in real time based on spectroscopy systems. Our detectors are based on sensors that are developed based on multi spectroscopy modes. See our website for more details.
We seek a Signal Processing , machine learning Engineer / Chemometrician who is multifaceted, creative, diligent in their work, and able to adapt to our dynamic environment. As part of a small multidisciplinary team, you will wear many hats, but your primary role will be to develop algorithms to interpret the signals from our array of chemical sensors. You will work closely with other SafetySpect employees and our R&D collaborators at Queens University Belfast.
Belfast, London or potential for virtual working (UK only)
- Develop algorithms to convert raw sensor signals into actionable data for our customers, including but not limited to data cleaning, detection, identification, and quantitation.
- Understand how basic principles of spectroscopy operation and how spectrometers are generated in response to biochemicals; relating artifacts in the data to the physical world is critical.
• Work with customers to define specifications for data outputs.
• Work with engineering team to integrate algorithms into prototypes and products.
• Lead data analysis effort. Review sensor data for quality and identify any anomalous behavior or artifacts that affect the performance of the algorithms.
• Build and maintain analytical tools used by members of SafetySpect's research staff and customers.
- B.S. degree or higher in Electrical Engineering, Data Science, Applied Mathematics, Statistics, Computer Science, Physics, or related field from an accredited college or university with 5+ years of experience.
- Experience in signal processing techniques.
- Experience with multivariate regression and classification methods, including but not limited to Principal Component Analysis (PCA), Partial Least Squares (PLS), adaptive cosine estimation (ACE), etc.
- Ability to generate operationally-relevant statistics for detection (e.g., probability of detecting a chemical at a given concentration), identification (e.g., probability of a correct identification, identification truth tables), and quantification (e.g., root mean square error, linear dynamic range).
- Ability to adapt data to compensate for non-idealities from instrumentation
- Proficiency in Python and related packages (scikit learn, scipy, numpy, pandas, matplotlib, etc.).
- Experience with developing algorithms that interface with hardware.
- Strong problem solving skills.
- Ability to work on an interdisciplinary team.
- Excellent written and verbal communication and interpersonal skills, including data presentation and a collaborative attitude.
- Eagerness to learn outside of your established field, i.e., learning chemistry, physics, etc.
- M.S. or Ph. D. in Electrical Engineering, Data Science, Applied Mathematics, Statistics, Computer Science, Physics, or related field from an accredited college or university with 5+ years of experience.
- Experience with multispectral imaging, chromatography, spectroscopy, or related hardware. Experience with sensors is of particular interest.
- Basic understanding of chemistry, particularly physical and analytical chemistry.
- Experience with dynamic modelling is a plus; dynamic models involving physical chemistry are especially relevant.
- Experience in multidimensional data visualization.
- Experience with machine learning and artificial intelligence, including data pipelines, cross-validation, and testing of models' robustness; knowledge of methods including random forests, convolutional and recurrent neural networks, k-nearest neighbors, etc.
Full-time for 6-month duration associated with a time defined, Innovate UK funded R&D program grant.
Potential pathway to full time, long term employment
Project Duration: May - October 2021