Vincent·Oct 8We benchmarked Mistral Large 4 and 13 other LLMs inside our platform.Nine reproducible test suites, fourteen models served from European data centers, deterministic graders and cross-checked LLM judges. What…
Vincent·Oct 4Your RAG Pipeline Is Leaking Personal Data. Here’s How We Stopped It with PresidioReversible PII masking in front of every LLM call: architecture, code, and a spaCy benchmark where the smallest model wins
Vincent·May 25, 2025Connecting Google Bigtable to FastAPI with async APIDiscover the async Bigtable interface for optimal performance.
Vincent·Jan 5, 2025Promptfoo: the secret to deploying reliable, secure artificial intelligenceIn the field of artificial intelligence, particularly in developing “intelligent” applications or conversational agents, the quality and…
Vincent·Dec 15, 2024MCP (Model Context Protocol): The Open-Source Protocol Revolutionizing AI IntegrationDiscover how MCP simplifies the integration of AI models into a data and tools ecosystem with a standardized and flexible solution.
Vincent·Oct 27, 2024Retriever Superpower with OpenAI Structured OutputRecently, I introduced the concept of structured outputs, also known as Structured Output, provided by OpenAI. Today, we’ll take things a…
Vincent·Oct 20, 2024It all started with a PerceptronIn homage to John Hopfield and Geoffrey Hilton, Nobel Prize winners for their “fundamental discoveries and inventions that made machine…
Vincent·Sep 29, 2024Building a Naive RAG from scratch with LlamaIndex, OpenAI, and ChromaAs AI continues to revolutionize various sectors, integrating knowledge retrieval with language models to build powerful applications is…
Vincent·Sep 21, 2024Unlock the Power of Structured Output with GPT-4oRevolutionizing the Integration of Generative Models into Smart Applications