<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>BigAI.vn — Engineering notes &amp; analysis</title><description>Articles on big data, machine learning, generative AI and MLOps from the BigAI engineering team.</description><link>https://bigai.vn/</link><language>en</language><item><title>What is a data lakehouse and how does it differ from a warehouse?</title><link>https://bigai.vn/en/blog/what-is-a-data-lakehouse/</link><guid isPermaLink="true">https://bigai.vn/en/blog/what-is-a-data-lakehouse/</guid><description>Data warehouse, data lake and lakehouse compared across six practical criteria, with a framework for choosing the right architecture at your scale.</description><pubDate>Wed, 12 Aug 2026 00:00:00 GMT</pubDate><category>Data architecture</category></item><item><title>How RAG works and when to fine-tune instead</title><link>https://bigai.vn/en/blog/how-rag-works/</link><guid isPermaLink="true">https://bigai.vn/en/blog/how-rag-works/</guid><description>Cost, accuracy and control compared between RAG and fine-tuning, with a practical decision framework for enterprises deploying internal AI assistants.</description><pubDate>Wed, 05 Aug 2026 00:00:00 GMT</pubDate><category>Generative AI</category></item><item><title>Cutting Spark cluster cost by 60%</title><link>https://bigai.vn/en/blog/spark-cost-optimisation/</link><guid isPermaLink="true">https://bigai.vn/en/blog/spark-cost-optimisation/</guid><description>Seven Spark optimisations applied on a real project, with before and after numbers — from storage format to partition sizing and executor memory.</description><pubDate>Tue, 28 Jul 2026 00:00:00 GMT</pubDate><category>MLOps</category></item><item><title>CDC with Debezium: core data onto the lakehouse</title><link>https://bigai.vn/en/blog/cdc-with-debezium/</link><guid isPermaLink="true">https://bigai.vn/en/blog/cdc-with-debezium/</guid><description>A practical Debezium setup for core banking and ERP systems, the mistakes that bite in production, and how to handle source schema changes mid-flight.</description><pubDate>Tue, 14 Jul 2026 00:00:00 GMT</pubDate><category>Data architecture</category></item><item><title>Evaluating an enterprise chatbot</title><link>https://bigai.vn/en/blog/evaluating-enterprise-chatbots/</link><guid isPermaLink="true">https://bigai.vn/en/blog/evaluating-enterprise-chatbots/</guid><description>The acceptance criteria BigAI uses for RAG systems: citation accuracy, correct refusal rate, document coverage and real user satisfaction.</description><pubDate>Tue, 07 Jul 2026 00:00:00 GMT</pubDate><category>Generative AI</category></item><item><title>Model drift: catch it before the business does</title><link>https://bigai.vn/en/blog/model-drift/</link><guid isPermaLink="true">https://bigai.vn/en/blog/model-drift/</guid><description>Which signals to monitor for production models, how to set alert thresholds that people will not ignore, and how to tell data drift from concept drift.</description><pubDate>Tue, 30 Jun 2026 00:00:00 GMT</pubDate><category>MLOps</category></item></channel></rss>