Sep 2022-present. AI Scientist II at the National Irish Center for Applied AI. I am part of the core team working on the Artificial Intelligence of (ICOS), an innovative IoT to Cloud Operating System supported by the European Union’s HORIZON research. My role involves developing AI APIs, working with High Performance Computer Centers (HPC), and implementing DevOps practices such as Docker, Kubernetes, singularity, and SLURM. I specialize in implementing cutting-edge generative models across various domains like Natural Language Processing (NLP), Federated Learning, time-series analysis, and computer vision.
Aug 2022-present. Computer Vision Research Fellow at Harvard University. Develop/assess Computer Vision and Deep Learning algorithms on imaging-based single-cell methods projects on CyCIF using high-dimension microscopy images for the Harvard’s Visual Computing Group and Harvard Medical School’s Lab of Systems Pharmacology. Methods include instance segmentation algorithms, unsupervised learning, image translation models and unbiased pixel-level feature extraction over Lateral Spillover in Segmented Images of Tissue with Pixel-level Cell State. Supervised by Hanspeter Pfister and Siyu Huang. Harvard john A. Paulson School of Engineering and Applied sciences.
Feb 2022-Aug 2022: Data Science intern at Harvard University: I worked at Harvard’s Hanspeter Pfister Lab improving instance segmentation and deep clustering models for cancer cell discovery using the FASRC Computing Cluster (HPC). I was priviledged to work with Siyu Huang, Edward Novikov, Hanspeter Pfister, amongst many other great scientists. Harvard john A. Paulson School of Engineering and Applied sciences.
Jan 2022-August 2022. Machine learning intern in EcoVadis. I developed a logo detection and verification framework based on the current state-of-the-art Object detectors, attribution and verification based on Contrastive text-image training, transfer-learning using transformers and image processing approaches. Under the supervision of Sophia Katrenko. Paris, France.
May 2020-November 2020. Computer vision and Artificial Intelligence Engineer in AI India Innovation centre. Developed the project “Land Use Land Cover Classification using advanced ML and DL techniques for Larger Study Area” project. The main tasks included accuracy assessment, creation of satellite imagery dataset from scratch, Training Models, Testing, Geospatial data and Cloud services, GPU, E2E cloud services and Colab. Valencia (Spain).
2021. Web/mobile developer for Ozone generator. COVID-19 Project Award for US$5200 for the creation of an Ozone generation, a web and mobile application, IEEE SIGHTS sights grant. Colombia.
2020. Web/mobile developer for mechanical ventilator Development of a mobile application for a mechanical ventilator project, awarded with a 3950 USD stipend. IEEE SIGHTS sights grant, Colombia.
2019. Electronic and Physics teacher. Teach and supervise Air-based flight machines, solar dials, lamps, flashlights, and programable robots to 400 teenagers aged between 4 and 16 years old in the United States of America. Camp Poyntelle, Pennsylvania, (U.S.A).
2016-2018. IT network technician. University of Cauca, Colombia. In charge of documenting all systems and network adjustments, as well as ensuring the general maintenance and hardware/software installation of telecommunication and network devices, at the different faculties of the University of Cauca. Colombia
2018. Computer science teacher. Developed Web Development and mobile applications as part of the general literacy curriculum program for Refugees, Social Hackers Academy, Athens (Greece)
2018. Front-end developer. Developed official web page designer at a psychiatric centre in Greece. Althaia psychiatric centre, Athens (Greece).
Oct 2013-present. General director & CEO. Astronomy Interpreter, public speaker, as well as in charge of the financial resources and logistics. Astronomical Observatory Francisco José de Caldas, Colombia.
Education
2020-2022. MsC. Joint Master on Image Processing and Computer vision [IPCV][News]. Full scholarship from Erasmus Mundus Joint Master (EMJM). The University of Bordeaux, the Autonomous University of Madrid, Peter’s catholic University of Pazmany. Location: Spain - France - Hungary, Scoring: 80%.
2020. Data Scientist - Correlation One. DS4A/Colombia 4.0. Data Science program organized by Colombian governement and Correlation One to learn from the world’s foremost experts to apply AI to real business initiatives at leading technology companies. We developped a prescriptive application with machine learning methods, AWS, and SQL technologies. Omnivida - MiniTIC
2014-2019. B.A. Electronics and Telecommunications engineering. University of Cauca. RANKING: 3° Position. Scoring: 84.488% GPA: 3.56
Grants, Awards and Honors
2022. Hack2hack first place biodiversity winner [Link]: Winner for the project “EvolveNet: A Meta-Attention Recommendation mechanism for seaturtle risk prediction and policy recommendation to minimize sea turtles endangerment anywhere on earth”.
2021. OpenCV 2021 competition phase 1 winner. Finalist in the OpenCV AI Competition 2021 sponsored by Microsoft Azure and Intel, the World’s biggest spatial AI competition. My team’s proposal was chosen from over 1400 submissions around the world, earning 6 OAK-D devices.
2021. Best president in the Aerospace & Electronics systems society Colombia. Prize of #AESS best President 2020 in Colombia Aerospace & Electronics Systems Society chapters. AESS Unicauca was also awarded as the best student chapters in Colombia. [Link]
2021. Awarded project (€5000 credit) for sponsoring request on Dengue forecasting : Towards a Smart Eco-epidemiological Model of Dengue in Colombia using Satellite in Collaboration with MIT Critical Data Colombia. Project supported by ESA Network of Resources Initiative. [Project]
2020. Embry Riddle Msc in Aerospace Eng. Scholarship - USA. Graduate Teaching Assistantship (GTA) in the Advanced Dynamics and Control Lab. Projects sponsored by the National Aeronautics and Space Administration (NASA) & SBIR/STTR Technologies.
2020. Hult-Prize regional awarded finalist in Popayan and finalist in Monterrey, Mexico in 2020.
2020. First place winner in IEEE R9 Humanitarian Activities for Ozone Purifier project (US $5,200). he IEEE R9 Humanitarian Activities Committee presented the COVID-19 Project Award for US$5200 for the creation of a Ozone generation, a web and mobile application. Co-authorship with Giovanna Ramirez, Germán Cambuya, Santiago Chicangana, Camilo Segura and Jesús Gurrute. Colombia
2020. First prize in IEEE SIGHTS covid-19 for development of mechanical ventilator (3914 USD award) [Link]
2020. DS4A data science selection: One of the 550 people selected for DS4A / Colombia over 12,000 applicants for advancing the AI knowledge and career. DS4A is an opportunity to learn the latest Data Science & AI techniques from a tenured Harvard professor through real-world business cases organized by #MinTIC and #CorrelationOne.
2020. Best impact global award solutions for COVID-19 in AAPM hackathon.Link
2019. Excellence award in the Regional Huawei ICT competition Colombia through Global Training and Certification department, Bogotá, Colombia S.A.S.
2016-2017. Half Honorific Scholarship at B.S. in the University of Cauca
2008-2013. Merit Scholarship awarded by Technical Comfacauca secondary school.
Thesis works
Logo verification and characterization with machine learning for document analysis. EcoVadis - Paris, France.
Mechanism for the characterization of motion artifacts in photophethysmography signals under low-intensity movements for tachycardia and bradycardia eventsTachycardia and Bradycardia Detection using Wearable Photoplethysmography under low-intensity Motion Artifacts. University of Cauca - Popayan, Colombia.
Publications
2024. Cajas, S. A., Restrepo, D., Moukheiber, D., Kuo, K. T., Wu, C., Garcia Chicangana, D. S., Paddo, A. R., Moukheiber, M., Moukheiber, L., Moukheiber, S., Purkayastha, S., Lopez, D. M., Kuo, P., & Celi, L. A. (2024). A Multi-Modal Satellite Imagery Dataset for Public Health Analysis in Colombia (version 1.0.0). PhysioNet. [online][code]
2024. DengueNet: Dengue Prediction using Spatiotemporal Satellite Imagery for Resource-Limited Countries. Kuan Ting Kuo, Dana Moukheiber, Sebastian Andres Cajas Ordonez, David Restrepo, Atika Rahman Paddo, Tsung-Yu Chen, Lama Moukheiber, Mira Moukheiber, Sulaiman Moukheiber, Saptarshi Purkayastha, Po-Chih Kuo, and Leo Celi [online]
2023. A. L. Suárez-Cetrulo, S. A. Cajas, J. Samanta, and R. S. Carbajo, “Machine Learning under Drifts and Shifts,” in Proceedings of the International Conference on Machine Learning, ISBN: 978-1-83508-358-1, 2023.[soon!]
2023. RSC Andres L. Suarez-Cetrulo, Sebastian Andres Cajas, Jaydeep Samanta. “Intelligence Layer - ICOS project Architecture White Paper”, [online]
2023. A comprenhensive review of Machine Learning at the Edge. S. A. Cajas Ordóñez, J. Samanta,A. L. Suárez-Cetrulo [soon!]
2023. S3 -TTA: Scale-Style Selection for Test-Time Augmentation in Biomedical Image Segmentation. Kangxian Xie, S. Huang, Hanspeter Pfister, Donglai Wei, S. A. Cajas Ordóñez [online]
2023. “G. J. Baker, E. Novikov, Y.-A. Chen, C. B. Hug, S. A. Cajas Ordóñez, S. Huang, C. Yapp, S. Coy, H. Pfister, A. Sokolov, and P. K. Sorger, “Contextual Cell State Inference in Whole-slide Multiplex Images of Tissue with Deep Convolutional Neural Networks,” IEEE Transactions on Medical Imaging, vol. 40, no. 7, pp. 1908-1919, Jul. 2021.” [on-review]
2021. “ECG Arrhythmia Classification Using Non-Linear Features and Convolutional Neural Networks. [online]
2020. “Modeling of motion artifacts on PPG signals for heart-monitoring using wearable devices”, Proc. SPIE 11330, 15th International Symposium on Medical Information Processing and Analysis, 1133014 (3 January 2020). [online]
2020. “Colcart - Web and mobile application to market and produce organic products in the department of Cauca”.[online]
2020. “Design and implementation of a satellite model for the Ibero-American Canned Satellite Competition”. [online]
For-fun open-source projects
Satellite Imagery Metadata Generator:MetaDengue is a unified dataset format that combines satellite imagery and Socioeconomical and environmental metadata. (Repository)
Modeling Motion Artifacts on wearable PPG signals for ambulatory heart rate monitoring. (Repository)
Vector-borne-satellite-predictor: Deep learning for vector-borne diseases with satellite-imagery based on ResNet50 and Vision Transformer. (Repository)
NASA Space Apps tutorials: A series of tutorials to download, integrate and apply machine learning and Deep learning to Satellite imagery. (Repository)
Multi-building tracker: Multi-target building tracker for satellite images using deep learning. A system for multi-target building tracking using satellite images has been developed.(Repository)
Generator & Multifunctional ozonator for the desinfection of viruses caused by COVID-19. A real-time air monitoring and purification system that allows the recording of multiple air variables through a mobile application that stores them remotely in the IBM Cloud service. The TLR5 phase design of a low-cost, portable, pre commercial ozonator equipment is also proposed as a vital device in the face of contamination caused by the COVID-19 virus in Colombia. Sponsored by IEEE Humanitarian Sights. (website) (Repository)
NASA image satellite analysis + Mechanical ventilator & Mobile App/website project: Design of Mechanical Ventilator, website and Monitoring Mobile application. Project Sponsored by IEEE Humanitarian Sights (Website) (Repository) (Mobile App)
2023. “Unlocking the Potential of Machine Learning at the Edge: Challenges and Future Trends”. By CeADAR: Sebastián Andrés Cajas, Jaydeep Samanta, Andrés L. Suárez-Cetrulo, Ricardo Simón Carbajo.(Online)
2021. Satellite imagery for NASA Space Apps challenge 2021. Conference with AESS Unicauca. April 23-25. (Conference)
Volunteering and open-source projects
2023. NASA SpaceApps lead: Leading the NASA SpaceApps 2023 that will take place in Colombia in October this year. (Event Website 2023)
2023. MIT Mentor: MIT Health Datathon 2023. (Event)
2021-present. Advisory Board member, Ex-President, Founder. AESS Unicauca, University of Cauca.
2020-present. Data Science Collaborator at MIT: PI Leo Celi. Collaborations at MITCriticalData-Colombia | 2020-2024.
2020. Computer vision researcher (Open-source): Smart Indoor Positioning, Guidance & navigation prototype system for the visually impaired. See our project here.
2019. Telecommunications engineer and team lead (Open-source) for can satellite telecommunications system in UAM challenge. Details here and simulation here.
2016. Italian teacher assistant for refugees: I worked as TA for Refugees from the Middle East and Africa on project supported by the European Union. AIESEC in Tortona, Italy
Professional Organizations
Aerospace & Electronic Systems Society founder & Advisor at University of Cauca(AESS-Unicauca)
IEEE
AIESEC
Red de Astronomia de Colombia - miembro (RAC)
Computer skills
+8 years experience with python and c++.
+5 years experience on java, SQL, typescript and shell.
+5 years working with Docker, Docker-compose and SLURM.
Focused on architecturing pipelines and real-time inference run-times (PyTorch,TensorFlow/Keras, TensorRT, Python, c++) on ML DevOps projects, in unsupervised or Self-supervised Learning applying Deep Learning to 3D Computer Vision problems.
I have experice with AWS, GCP, FASRC Harvard cluster, E2E, Microsoft Azure and private servers (University of Bordeaux (UbX), Autonomous University of Madrid (UAM)).
Front/back-end experience with web/mobile development (React, Flutter, Ionic, Ruby) & Cloud experience with AWS, Firebase, IBMCloud, MongoDB, Microsoft Azure, E2E, GCP)
Research Interests
- I am interested in the multiple applications of Reinforcement Learning, Data Science, Computer Vision, machine learning models, and Space Sciences (Aerospace, radioastronomy, rocketry, satellite engineering). And I do research for fun and the sake of learning from amazing people around the world.