Data Scientist

SBQuantum•Sherbrooke, QC

2d•Hybrid

About The Position

SBQuantum is a seed-stage startup at the forefront of magnetic sensing innovation. Our vision is to unlock the full potential of magnetic intelligence through our proprietary quantum diamond technology and curated algorithms. By combining diamond-based quantum sensors with advanced AI-driven software, we transform complex magnetic field data into actionable, high-value insights. Our solutions enable accurate magnetic mapping, navigation and object detection in environments where traditional sensing technologies—like GPS, radar, imagery, or sonar—cannot perform. From public safety and defence to space exploration, our multidisciplinary team of engineers, physicists, and data scientists is redefining how the world perceives magnetic signals—turning invisible complexity into clear, useful intelligence.

Requirements

5 years of experience in a startup environment
Background in magnetic compensation algorithms, ML processing pipelines and data science
Experience in autonomous platforms deployment
Drive to engage with prospective, current clients and engage at conferences to disseminate product knowledge
Comfortable wearing multiple hats and switching contexts quickly
Strong problem-solver with a bias toward action
Excellent communication skills (written and verbal)
Experience with custom Python code

Nice To Haves

Experience in deeptech, hardware, or scientific environments
Bilingual English/French

Responsibilities

Design, implement, and validate vector magnetic compensation algorithms; including Tolles-Lawson and extended models to characterise and remove platform-induced magnetic interference across all three field components. Adopt a rigorous testing against ground truth and in-flight datasets, and an awareness of how compensation quality directly impacts end-user navigation performance.
Interrogate large, multi-channel magnetic and inertial datasets to identify systematic patterns, interference signatures, and anomalous behaviour; applying statistical analysis, spectral methods, and machine learning techniques to extract actionable insight from complex, noisy signals in operational navigation contexts.
Build and maintain structured test frameworks to benchmark compensation performance across platforms, flight regimes, and environmental conditions; tracking residual error metrics, iterating on model parameters, and documenting improvement cycles with reproducible results that can be clearly communicated to navigation system integrators.
Develop robust, well-documented Python pipelines for ingesting, synchronising, and pre-processing multi-sensor data streams — ensuring consistent data formats, calibration traceability, and version control across field campaigns and laboratory experiments, with outputs structured to meet the ingestion requirements of downstream navigation systems.
Work closely with geophysicists, INS/navigation engineers, and platform specialists to align compensation outputs with navigation requirements; engaging directly with end users in the navigation space to understand operational constraints, gather feedback on delivered data products, and ensure algorithm development remains grounded in real-world mission needs.
Critically assess the validity of compensation models and their underlying assumptions across varying operational contexts; challenging results that appear too clean, identifying failure modes under edge-case conditions, and proposing alternative modelling approaches where standard methods reach their limits, with findings fed back to relevant stakeholders and end users.
Communicate findings clearly to both technical and non-technical stakeholders; including prospective and active end users in the navigation domain. Produce well-structured reports, visualisations, and presentations that distill complex compensation performance results into clear conclusions informing system design decisions, procurement discussions, and operational planning.