Ideas, builds & research
Practical notes on blockchain, applied AI, cybersecurity and web development.
Generative Engine Optimization: how to tell whether an AI can cite your website
Ranking signals and citation signals are not the same thing. A model does not reward your domain authority, it rewards content it can extract, attribute and verify. Here are the four layers that decide it, and a free way to measure yours.
How to choose a technology company in Andalusia
Before comparing proposals, check the company exists and can be verified. Then ask about subcontracting, code ownership, testing and what happens the day they leave. A practical guide, with the public sources you can check yourself.
Smart contract development: what changes when the code holds the money
Writing a smart contract is not writing a backend. The code is immutable, its state is public and every user is a potential attacker. Here is how we build them, which chain we pick and what "finished" actually means.
Basalt: an open-source purple-team of AI agents that proves vulnerabilities by execution
We built Basalt, an open-source security tool where AI agents do not just guess at bugs — they prove each finding by reproducing its exploit in an isolated sandbox. Here is what it does and why it matters.
What is a self-learning RAG system and why it matters for your business
Retrieval-Augmented Generation lets an AI answer with your own data — and a self-learning loop makes it sharper every day. Here is how it works and where it pays off.
Smart contract audits: why they are non-negotiable before launch
On-chain code is immutable and handles real value. A single overlooked bug can drain a protocol. Here is what a serious audit covers and why it pays for itself.
Carbon credit offsetting with blockchain: traceable, verifiable, trusted
Voluntary carbon markets have a trust problem: double-counting and opaque registries. Blockchain makes every credit traceable from issuance to retirement.
Private AI: use LLMs without handing over your data
You do not have to choose between AI capability and data confidentiality. Private AI brings the model to your data, not your data to someone else.
AI applied to tourism: from search to unforgettable experiences
Travelers do not want a list of 1,000 options — they want the right one. AI turns a huge catalogue into a personal concierge.
Post-quantum cryptography: getting ready for Q-day
A large enough quantum computer will break the encryption that protects today’s internet. The threat is "harvest now, decrypt later" — and it has already begun.
End-to-end encrypted communications: what it really takes
Encryption is not a checkbox. True end-to-end security depends on key management, forward secrecy and a threat model that matches your reality.
AI-powered defensive systems: protecting critical infrastructure
Defensive AI is about detection, resilience and response — protecting the systems a society depends on, faster than human teams alone could.
IoNT — the Internet of Natural Things, explained
The Internet of Things connected our devices. The Internet of Natural Things asks a bolder question: what if the natural world itself became living infrastructure?
Blockchain beyond crypto: real uses for your business
Forget the token hype. The lasting value of blockchain is boring and powerful: shared, tamper-evident records that multiple parties can trust without a middleman.
AI for process automation: where it actually pays off
The biggest AI wins are rarely flashy. They are the repetitive, judgment-light tasks quietly draining your team’s hours every single day.
AI and privacy: building systems that respect data by design
Privacy is not the price of using AI. Done right, AI and data protection reinforce each other — and regulators are increasingly making it mandatory.