Background
David is a senior engineer who has spent the past decade as a Technical Lead at a major European multinational telecommunications company, where he designs and delivers AI-driven automation platforms, distributed enterprise systems, and cloud-native architectures at scale. His work spans LLM integration, multi-agent orchestration, retrieval-augmented generation, and the migration of large legacy codebases to modern cloud environments.
Before entering industry, he worked as a researcher at the University of Murcia (Spain), focused on artificial intelligence, semantic reasoning, and context-aware systems for ambient assisted living. His academic work produced multiple peer-reviewed publications in international journals and conferences.
Teaching by building
David has a consistent thread of making complex topics accessible through hands-on projects. He authored a complete tutorial teaching Java 8 through the development of a poker game — covering streams, lambdas, concurrency, and design patterns by building something real instead of abstract exercises. He published a companion guide on JavaHispano, one of the largest Spanish-speaking developer communities. The same teaching instinct drives the Pico Framework: a modular Python toolkit for AI agent systems designed so that others can learn by reading the source.
That instinct is also behind CleanSky’s learning hub and DeFi course — 80+ articles and guides translating complex DeFi mechanics into plain language across four languages.
What he built
CleanSky started as a personal need: a way to understand a multi-chain DeFi portfolio without connecting a wallet, creating an account, or trusting a third party with private data. No tool on the market offered that combination of depth, privacy, and clarity — so he built one.
The project is entirely independent. It has no venture capital, no affiliate revenue, and no sponsored content. Every article in the blog and learning hub is written to educate, not to sell. The editorial policy explains why.
CleanSky — Portfolio Intelligence
A read-only, privacy-first DeFi portfolio tracker covering 484+ protocols across 34+ networks. Java backend with multi-chain EVM indexing, React frontend. Protocol position aggregation and classification (Aave, Uniswap, Morpho, Lido, and hundreds more), risk visualization, underlying asset decomposition, and health factor tracking. Designed for mainstream DeFi adoption with banking-grade UX simplicity. No sign-up, no wallet connection, zero tracking.
CleanSky Contactless — NFC Crypto Payments
Open-source Android app for contactless cryptocurrency payments via NFC. Built with Kotlin and Jetpack Compose. Private keys never leave the device, secured via Android Keystore (AES-256-GCM) with hardware-backed security (StrongBox/HSM). Supports multiple EVM chains (Base, Polygon, Arbitrum, Optimism) and execution modes including meta-transactions (Gelato/Biconomy) and ERC-4337 Account Abstraction with Paymaster.
CMSP — Capped Mandate Subscription Protocol
An open-source payment primitive for secure, non-custodial recurring payments on-chain. Designed to replace infinite token approvals with cryptographically bounded mandates (max amount, period, expiration). Zero-custody architecture with formal specification (8 RFCs), threat model, and reference implementations across EVM (Solidity), Solana (Anchor), and Cosmos (CosmWasm). Proposed and discussed on Ethereum Magicians.
Pico Framework — AI Agent Platform
A modular open-source Python framework for building AI-enabled applications and agent-based systems. Multi-package ecosystem including dependency injection, lifecycle orchestration, plugin discovery, and agent execution infrastructure. Integrated with FastAPI, SQLAlchemy, Celery, and Pydantic. Supports protocol-based LLM orchestration, multi-provider model routing, and tool lifecycle governance. Published on PyPI.
Research & publications
- “Automatically Calibrated Occupancy Sensors for an Ambient Assisted Living System” — Integrated Computer-Aided Engineering, 2016.
- “An Energy-Efficient Middleware for Ad-hoc AAL Wireless Sensor Networks” — Elsevier Science, 2012.
- Multiple international conference papers (IWINAC, EUROCAST, IWAAL) on AI reasoning, context-awareness, and sensor-based automation.
Open-source contributions
Contributor to Spring AI, Spring Security, GNOME/Ubuntu, and the JavaHispano developer community. GitHub Arctic Code Vault Contributor. 47 public repositories spanning Java, Python, JavaScript, Kotlin, Solidity, and infrastructure tooling.
Education
- Master’s in AI and Advanced Telematics — University of Murcia, Spain (2009–2010)
- Degree in Computer Engineering (Software Management) — University of Murcia, Spain (2003–2007)