Jorge Cardete
Email: jorgecardete@gmail.com | LinkedIn: jorgecardetellamas
Experience
Bit2Me
Designing and running experiments across the activation, retention and acquisition funnels, and building the technical infrastructure behind growth (tracking, automation, personalization).
Grupo Cardete Ventures
Leading technology at a 24/7 urgent roadside assistance and express transport company operating in Madrid and Barcelona. Building the tech platform behind the operation: dispatch, logistics and customer experience for a service that reaches drivers in under 30 minutes.
The Deep Hub
Founded and run The Deep Hub, a technical publication on data science and machine learning hosted on Medium. Grew it to 570+ published articles and ~100 contributing engineers and researchers, with 2.2K followers.
I write in-depth technical pieces (mainly computer vision) and edit every submission to the publication.
Zapic
Building AI assistants for social media management: Zapic researches trends, generates content and engages with communities. I led product, engineering and company strategy.
xFractal
Led AI initiatives with on-chain AI agents inside Solana.
Designed and implemented solutions in Rust, optimizing AI models and orchestrating scalable infrastructure on the blockchain.
Axpe Consulting
BBVA: Developed an AI-powered chatbot to address banking regulatory issues, integrating AI agents on Cohere, GPT and Llama to provide accurate and efficient responses.
Dragados: Collaborated with a team of data scientists and analysts to create real-time model result visualizations in Grafana within the Databricks ecosystem.
Cosentino: Built a computer vision model using Azure ML to classify production outputs, optimizing processes and ensuring quality control.
Omya: Developed a monitoring and analytics system using SAS Viya and Python, providing end-to-end visibility and real-time tracking of company operations.
Sabadell: Created a testing framework for speech-to-text models, including Whisper, Wav2Vec and Kaldi, to evaluate their performance and accuracy in real-world applications.
Medtronic
Salesforce: Analyzed and managed data using Power BI and Python to generate actionable insights, improving business processes and decision-making.
OCR: Developed an OCR solution in Python to automate the extraction and processing of unstructured tender documentation.
FTT DAO
Skills
- Languages: Python, Rust, TypeScript, C++, SQL
- Machine learning: PyTorch, TensorFlow and Keras, OpenCV, LightGBM, Azure ML, CNNs, object detection, OCR, speech-to-text
- Generative AI: LLMs (GPT, Llama, Cohere) and AI agents, including on-chain on Solana
- Data: Databricks, Grafana, Power BI, SAS Viya, SQL
- Backend and infrastructure: Flask, Docker, Azure, Terraform, RabbitMQ, MySQL
- Web: Astro, TypeScript, HTML and CSS
- Growth: Funnel experimentation, tracking, automation, personalization, Google Ads and Meta Ads
Projects and hackathons
- Personal portfolio (2026): This site: a portfolio and blog built with Astro and Markdown content, custom pixel-art icons and a playable map to explore it.
- X-Ray · HackSpain (2026): Embat challenge: a 0–100 financial health score for SMEs built from their treasury data, with a 3-month forecast, alerts and a scenario simulator. Demo · Code
- Colosseum (2025): Top 10 at the Colosseum Solana hackathon with xFractal, a trading platform powered by AI agents. Video
- Home Improvement Scraper (2024): Real-time price comparison API across Leroy Merlin, BricoDepot and Bauhaus, with distributed workers and caching. Code
- Cell Vision (2024): Detection and classification of peripheral blood cells: OpenCV segmentation and a fine-tuned ConvNeXt (98.5% accuracy), served in a Flask app. Code · Model · Article
- Parking Detector (2024): Detects free parking spaces from video with a TensorFlow CNN, deployed with Docker and Azure. Code · Article · Deployment
- Telegram Ops Bot (2024): Telegram bot to run The Deep Hub: it finds writers on Medium, handles email submissions and generates social posts with AI, each task in its own container. Code · Social API
Writing and publications
- Blog: 52 articles on computer vision, machine learning, AI and Python on this blog.
- Research paper: How well do different chatbots respond to multiple myeloma treatment guidelines? Leukemia (Nature Portfolio, April 2025). Edwin Uriel Suárez, Fabio Torres-Saavedra, Amalia Domingo-González, Jorge Cardete, Pilar Llamas-Sillero.