Paul SerbanSoftware Engineer
  • PORTFOLIO
  • BLOG

#Evaluation

Posts

  • Command vs. query prompts: a CQRS-inspired framework for structuring LLM interactionsTreating action-oriented and retrieval-oriented prompts as fundamentally different concerns leads to cleaner, more predictable AI behaviour

    Apply CQRS thinking to prompt engineering — learn how splitting command and query prompts improves LLM reliability, tracing, and eval coverage.

    • #AI Engineering
    • #LLM Architecture
    • #Prompt Engineering
    • #CQRS
    • #System Design
  • RAG 101 for AI Engineers: From Naive Retrieval to Production-Grade PipelinesChunking, embeddings, reranking, citations, evaluation, and failure modes explained simply.

    A step-by-step guide to building a reliable RAG system, covering chunking, embeddings, retrieval, reranking, context windows, and evaluation tactics for better answers.

    • #AI Engineering
    • #RAG
    • #Retrieval-Augmented Generation
    • #LLMs
    • #Embeddings
View all posts
  • LinkedInLinkedIn
  • GitHubGitHub
  • HackerRankHackerRank
  • LeetCodeLeetCode
  • EmailEmail
  • Portfolio
  • My Projects
  • Coursework
  • Blog
  • Posts
  • Snippets
  • Book Notes
  • Cookie Settings
  • Cookie Policy
2026 © Paul Serban. All rights reserved.www.paulserban.eu | Sitemap