- 2011 · Kent
- 2016 · Middlesex
- 2018 · Polytechnics
- 2019 · ALU
- 2020 · Accenture
- 2023 · FJTechlab
Research that ships.
Systems that cite their sources.
Shivanand Guness is a data scientist and AI engineer with a Ph.D. in Electronic Engineering, building privacy-preserving AI in production and publishing the computer-vision and assistive-technology research it grew out of — sixteen peer-reviewed works, one working engineer. Every claim on this page resolves to a real reference16.
Abstract
Abstract. Recognised with the Morpho UK Award for Excellence in Biometrics, Shivanand Guness holds a Ph.D. in Electronic Engineering (University of Kent) built on computer-vision and assistive-technology research — head-tracking systems for wheelchair and switch control, published across conferences and an M.Sc. in Information Security and Biometrics (Distinction). That research background now runs in production: he designs and ships privacy-preserving AI systems†, most recently a PII-anonymisation microservice deployed under European data-sovereignty constraints at FJTechlab International. The two halves are not separate careers — the same rigor that gets a paper past peer review is what gets a system past an audit.
Experience
- 2023–Present
Senior Full Stack Engineer
FJTechlab International · Mauritius
- Designed and launched a privacy-enhanced AI document-extraction and PII-anonymisation microservice — FastAPI, Microsoft Presidio, spaCy — with custom entity recognisers, packaged with Docker for production.
- Evaluated and deployed self-hosted AI models on European cloud infrastructure (OVH) to meet data-sovereignty and privacy requirements.
- Led development of a multi-platform application across web and desktop clients.
- Built and maintained ETL/SSIS data pipelines with Power BI reporting, including cross-database reconciliation against production data-warehouse schemas.
- 2020–2023
Application Development Specialist
Accenture · Mauritius
- Developed and deployed applications in C# and C++, enhancing system functionality and user experience.
- Streamlined development processes with cross-functional teams, reducing deployment time.
- Led cross-platform integration efforts.
- 2019–2020
Academic Lead — AI & Innovation Programme
The African Leadership University · Mauritius
- Owned academic vision and strategy; revised curriculum to meet accreditation standards.
- Created and ran live/streamed hackathons every three months in Johannesburg, South Africa.
- Drove student retention to 95% through programme metrics and reporting.
- 2018–2019
Lecturer, Software Development & Engineering
Polytechnics Mauritius Ltd · Réduit, Mauritius
- Designed and delivered industry-aligned courses in software development and engineering.
- Built project-based assessments bridging theory and real-world application.
- 2016–2018
Research Assistant
Middlesex University Mauritius · Vacoas, Mauritius
- Developed research software tools that improved research efficiency by 20%.
- Contributed data analysis to three peer-reviewed publications.
- 2011–2014
Part-time Lecturer / Demonstrator, School of Engineering and Digital Arts
University of Kent · Canterbury, United Kingdom
- Conducted lessons on PHP, MATLAB, and C# to improve programming proficiency.
- Guided students in XML applications and project management practice.
Education
- Ph.D., Electronic EngineeringUniversity of Kent, United Kingdom
- M.Sc., Information Security and Biometrics (Distinction)University of Kent, United Kingdom
- Diploma in Technology, Software Analysis and Design (Very Good)Mauritius Chamber of Commerce and Industry
Certifications
- Microsoft Certified: Azure Data Scientist Associate (2019–2024)
- Microsoft Certified Solutions Associate: Machine Learning (2019)
- Microsoft Certified Professional
- Microsoft Certified Technology Specialist: SQL Server 2005 (2008)
- Microsoft Certified Solution Developer: .NET (2008)
- Microsoft Certified Application Developer: .NET (2006)
- Certified ScrumMaster
- Linux Professional Certification
Awards
- Morpho UK Ltd Award for Excellence in Biometrics
Selected work
Drawn from the roles and research above — not the sparse public-repo history below, which is coursework, not flagship work.
Grand Bassin Pilgrimage Datasets (Kanwar & Traffic Imagery)2026
Two anonymised imagery datasets from the Maha Shivaratri pilgrimage at Grand Bassin, Mauritius — pilgrim kanwar imagery and aerial/roadside traffic imagery — released publicly for research use.
Published as open datasets on Zenodo (see Publication List 1–2); a companion Q&A demo is deployed as a public Hugging Face Space.
- Computer Vision
- Dataset Anonymisation
- Streamlit
Privacy-Enhanced AI Document Extraction2023 — Present · FJTechlab International
A production PII-anonymisation microservice built on FastAPI, Microsoft Presidio, and spaCy, with custom entity recognisers and multiple anonymisation strategies, packaged with Docker.
Deployed for sensitive document handling under GDPR / European data-sovereignty constraints, with self-hosted models on OVH cloud infrastructure rather than third-party APIs.
- FastAPI
- Microsoft Presidio
- spaCy
- Docker
- OVH
Depth-Based Head Tracking for Assistive Control2011 — 2021 · PhD research, University of Kent
A multimodal real-time head-tracking and facial-gesture system evaluated as a mouse emulator and proportional wheelchair controller for users with neuro-disability.
Core PhD research thread, published across six peer-reviewed venues (see Publication List 3, 11–13); evaluated vision-based head-trackers as assistive-technology input devices.
- Computer Vision
- Depth Sensing
- Gesture Recognition
Low-Cost Air Quality & IoT Monitoring2018 — 2020 · Middlesex University Mauritius
Low-cost sensor networks for air-quality monitoring in Mauritius, plus AR-assisted disaggregated energy monitoring and IoT control platforms.
Contributed to threat-modeling for IoT security and a coding survey of cyber-physical systems challenges (Publication List 4–9).
- IoT
- Low-cost Sensors
- Threat Modeling
Code & demos
- github.com/shivam11 — more repos are being made public over time; the profile is the current source of truth.
- huggingface.co/spaces/shivam11/budgetQnA — a public Streamlit Space. Full profile: huggingface.co/shivam11.
Publication List
Sixteen peer-reviewed works, 2011–2026. Two conference entries carry no DOI on record and are listed without a link — nothing here is a placeholder.
- [1]Guness, S. P. (2026). Grand Bassin Kanwar Dataset: Anonymised Imagery of the Maha Shivaratri Pilgrimage, Mauritius (v1.0). Zenodo. doi.org/10.5281/zenodo.21321319
- [2]Guness, S. P. (2026). Grand Bassin Traffic Dataset: Anonymised Aerial and Roadside Traffic Imagery from the Maha Shivaratri Pilgrimage, Mauritius (v1.0). Zenodo. doi.org/10.5281/zenodo.21321448
- [3]Guness, S. P., Deravi, F., Sirlantzis, K., Pepper, M. G., & Sakel, M. (2021). Development and evaluation of a mouse emulator using multimodal real-time head tracking systems with facial gesture recognition as a switching mechanism. Lecture Notes in Computer Science. doi.org/10.1007/978-3-030-68790-8_18
- [4]Okigbo, C. A., Seeam, A., Guness, S. P., Bellekens, X., Bekaroo, G., & Ramsurrun, V. (2020). Low cost air quality monitoring. Proc. 2nd International Conference on Intelligent and Innovative Computing Applications. doi.org/10.1145/3415088.3415124
- [5]Seeam, A., Ogbeh, O. S., Guness, S., & Bellekens, X. (2019). Threat Modeling and Security Issues for the Internet of Things. 2nd International Conference on Next Generation Computing Applications (NextComp 2019). doi.org/10.1109/NEXTCOMP.2019.8883642
- [6]Purmaissur, J., Seeam, A., Guness, S., & Bellekens, X. (2019). Augmented Reality Intelligent Lighting Smart Spaces. 2019 Conference on Next Generation Computing Applications (NextComp). doi.org/10.1109/nextcomp.2019.8883577
- [7]Igbokwe, K., Seeam, A., Guness, S., & Towakel, P. (2019). Challenges Ahead in Cyber Physical Systems: A Coding Survey. 2019 Conference on Next Generation Computing Applications (NextComp). doi.org/10.1109/nextcomp.2019.8883644
- [8]Towakel, P., Guness, S. P., Seeam, A., Bekaroo, G., & Bellekens, X. A. (2018). Air quality monitoring in Mauritius. 2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC 2018). doi.org/10.1109/ICONIC.2018.8601245
- [9]Purmaissur, J. A., Towakel, P., Guness, S. P., Seeam, A., & Bellekens, X. A. (2018). Augmented-reality computer-vision assisted disaggregated energy monitoring and IoT control platform. ICONIC 2018. doi.org/10.1109/ICONIC.2018.8601199
- [10]Omoogun, M., Ramsurrun, V., Guness, S., Seeam, P., Bellekens, X., & Seeam, A. (2017). Critical patient eHealth monitoring system using wearable sensors. 1st International Conference on Next Generation Computing Applications (NextComp 2017). doi.org/10.1109/nextcomp.2017.8016194
- [11]Guness, S. P. (2015). A novel depth-based head tracking system for proportional wheelchair control. TechAAL 2015 (conference proceedings). no DOI on record
- [12]Guness, S. P., Deravi, F., Sirlantzis, K., Pepper, M. G., & Sakel, M. (2013). A novel depth-based head tracking and gesture recognition system. Assistive Technology Research Series. doi.org/10.3233/978-1-61499-304-9-1021
- [13]Guness, S. P., Deravi, F., Sirlantzis, K., Pepper, M. G., & Sakel, M. (2012). Evaluation of vision-based head-trackers for assistive devices. Proc. Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS). doi.org/10.1109/EMBC.2012.6347068
- [14]Guness, S. P. (2012). Poster 134: Developing a Vision Based Gesture Recognition System to Control Assistive Technology in Neuro-Disability. Archives of Physical Medicine and Rehabilitation. doi.org/10.1016/j.apmr.2012.08.172
- [15]Deravi, F., & Guness, S. P. (2011). Gaze trajectory as a biometric modality. BIOSIGNALS 2011 — Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing. no DOI on record
- [16]Guness, S. P. (2021). Contributor, Chapter 45: "Low-Cost Sensors and the Quality of Data". In Macey, T. (Ed.), 97 Things Every Data Engineer Should Know, O'Reilly Media. www.oreilly.com/library/view/97-things-every/9781492062400/ch45.html
Correspondence
Both addresses are encoded in the markup and only assembled into a real address in your browser on click — not left as plain text for a scraper to lift.