Clear definitions of the key concepts we use in euskolabs. Designed for customers, search engines and IA engines.
Digital engineering laboratory that combines web development, software and IA under the same roof. Unlike a traditional agency, it operates with surgical execution speed and dedicated technology partner attention.
Application cloud-hosted software, accessible via browser, where customers pay a regular subscription. It does not require local installation and is automatically updated.
Nicho SaaS focused on solving a specific problem for a small audience, with minimum equipment and low infrastructure costs. Example: Aegis Emergency Button.
Artificial intelligence system that perceives its environment (text, voice, images), information-related location and runs actions to meet objectives autonomously. You can integrate voice (Kokoro TTS, Whisper), vision and automation (n8n).
Discipline of collecting, analysing and structuring information from publicly accessible sources. In context B2B: data extraction from public directories, Google Maps and corporate websites for commercial prospecting.
Technique that combines information recovery from a knowledge base with text generation by a LLM. It allows more accurate and up-to-date responses, grounding to the model in verifiable data.
A model trained with vast amounts of text to understand and generate natural language. Examples: GPT, Gemini, LLAMA. In euskolabs are used in both cloud and local mode (Olama).
Content optimization for IA engines such as ChatGPT, Gemini and Perplexity. As the SEO places the content (schema markup, answer-first format, FAQ) on Google, the AEO structure for LLMs to quote as an authoritative response.
Specific optimization for generative engines that produce synthesized responses from multiple sources. AEO subset focused on Google AI Overviews and SGE.
Software architecture where a single instance of the application serves multiple (tenants) customers with their completely isolated data. It allows shared and scalated costs without additional deployments.
Minimum version of a product with sufficient functionalities to validate the value proposal and monetize. In Basque, the SaaS MVP are delivered in less than 14 days.
JavaScript API to render 3D and 2D graphics in the browser without plugins. euskolabs uses WebGL for 3D animations, particles and visual effects on premium websites.
Open source natural voice synthesis engine used by euskolabs for AI Agents with voice. Low latency, multi-language support, natural quality.
Optimized version of Whisper (OpenAI) for real-time voice recognition. Used in euskolabs to make the AI Agents understand user voice with minimum latency.
Open source workflow automation platform. euskolabs uses it to connect AI Agents to business systems: CRM, ERP, email, SMS, databases.
Alternative open-source to Firebase based on PostgreSQL. Provides database, authentication, storage and real-time. Used by euskolabs as a multi-tenant SaaS backend.
Python framework to build high-performance APIs. Automatic documentation, native async, type validation. Main backend of euskolabs for SaaS and AI Agents.