Juan Felipe Jaramillo Hernández, M.Sc.

Aimira — an AI platform for real-time textile quality control, detecting fabric defects on rolling machines directly on the factory floor.

• ▹Lead the computer-vision direction of the product — owning the major design choices and the direction of the deep-learning algorithms behind real-time defect detection.
• ▹Architect and write the core of the system — the vision/DL components and the on-device application monitoring rolling machines live in production.
• ▹Direct AI agents and serve as the human in the loop for AI-assisted development, sustaining engineering velocity without compromising reliability.
• ▹In production across 10+ industrial deployments: 2M+ defects detected over 52K+ inspected fabric batches.

AI Developer — Computer Vision

• ▹Designed and deployed state-of-the-art vision algorithms for industrial quality inspection, reaching >90% defect-detection accuracy and validating the system in production (TRL 4–7).
• ▹Built dataset-generation and annotation pipelines guaranteeing model precision and reproducibility across product lines.
• ▹Developed optimization and ML methods for automated resource management, improving efficiency and cutting costs in event-logistics planning.
• ▹Scouted and evaluated emerging AI techniques, surfacing new R&D&i opportunities and strengthening the company's innovation capabilities.

Intern — Technology Consulting, Digital & Emerging Tech

• ▹Automated internal client processes with RPA, reducing manual effort by 30%.
• ▹Designed and implemented APIs for a blockchain-based traceability solution, improving data integrity and transparency.
• ▹Conceptualized a multimodal CV/ML platform to optimize logistics and stock distribution for brick-and-mortar retail.

Research Assistant — Precision Agriculture

• ▹Researched sustainable agricultural practices through multiscale optimization within ÓMICAS Macroproject 4.
• ▹Created the first plant light-field image dataset, pioneering light-field research in agricultural imaging.
• ▹Developed and evaluated a spatial super-resolution deep-learning model, improving image quality for precision farming.

Robotics & Society — RAS/IEEE · Social Practicum

• ▹Designed, built and programmed sumo-robots (sensors, actuators, microcontrollers); competed at Runibot 2018.
• ▹Delivered multimedia programming tutorials to underserved youth, impacting 50+ students and fostering digital literacy with Scratch.

PhD in Computer Science · Industrial Mention

Advanced machine-learning algorithms for computer vision applied to sustainable agriculture, at the Valencian Research Institute for AI (VRAIN). Industrial agreement with Arisnova on the VisionTex innovation project — an AI quality-inspection system for industry.

M.Sc. Artificial Intelligence, Pattern Recognition & Digital Imaging

GPA 8.9 / 10. Specialization in ML for Computer Vision, Automatic Speech Recognition and NLP. Thesis: a novel object-detection method with integrated depth estimation for real-time systems.

B.Sc. Electronic Engineering · ABET-accredited

GPA 4.2 / 5.0. Foundations in control systems, telecommunications, digital systems and electronic circuits. Thesis: deep-learning spatial super-resolution of plenoptic plant images (ÓMICAS).

Application of Machine Vision Techniques in Low-Cost Devices to Improve Efficiency in Precision Farming. Jaramillo-Hernández, J.F., Julian, V., Marco-Detchart, C., & Rincón, J.A. Sensors 24(3): 937. doi.org/10.3390/s24030937 ↗

Efficient Depth Object Detection: Ablation-Driven Optimization for Lightweight YOLOv8 Architecture. Jaramillo-Hernández, J.F., Julian, V., Marco-Detchart, C., & Rincón, J.A. PAAMS 2024, Communications in Computer and Information Science, vol. 2149, Springer, Cham. doi.org/10.1007/978-3-031-73058-0_13 ↗

Oral presentation, 22nd International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS'24) — "Efficient Depth Object Detection".