[{"content":"Hey I\u0026rsquo;m Hicham Assoudi.\nI help Oracle practitioners, architects, developers, dbas, and enterprises make better AI implementation decisions by combining Oracle depth, real experiments, and research-grade evaluation.\nI’m an Oracle consultant, AI researcher, and technical author.\nMy work sits at the intersection of enterprise Oracle technologies and practical artificial intelligence. I focus on topics such as Oracle Database AI, OCI Data Science, multilingual AI, and enterprise research-driven AI implementation patterns.\nOver the course of my career, I have built deep experience in Oracle technologies across architecture, integration, data platforms, and enterprise delivery. In parallel, my academic path led me to a PhD in Artificial Intelligence and Natural Language Processing, with a long-standing interest in multilingual AI and domain-specific NLP.\nI created assoudi.blog as my personal knowledge hub.\nThis is where I publish practical, technically grounded writing for Oracle practitioners, architects, developers, and technical leaders who want to go beyond generic AI discourse and understand how Oracle and AI can be combined in a useful, disciplined, and enterprise-relevant way.\nHere I write about:\nOracle AI, vector search, RAG, and Agents In-DB semantic search and retrieval patterns using ONNX Models Oracle OCI Data Science (training models from scratch, fine-tuning models, etc.) architecture trade-offs and implementation decisions model evaluation and benchmarking enterprise AI patterns for Oracle environments This site also connects my articles, books, GitHub repositories, Hugging Face assets, videos, and selected technical resources.\ndisclaimer: I speak for me not my employer\n","permalink":"https://assoudi.blog/about/","summary":"\u003cp\u003eHey I\u0026rsquo;m \u003cstrong\u003eHicham Assoudi\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eI help Oracle practitioners, architects, developers, dbas, and enterprises make better AI implementation decisions by combining Oracle depth, real experiments, and research-grade evaluation.\u003c/p\u003e\n\u003cp\u003eI’m an Oracle consultant, AI researcher, and technical author.\u003c/p\u003e\n\u003cp\u003eMy work sits at the intersection of enterprise Oracle technologies and practical artificial intelligence. I focus on topics such as Oracle Database AI, OCI Data Science, multilingual AI, and enterprise research-driven AI implementation patterns.\u003c/p\u003e","title":"About"},{"content":"This page gathers my books and long-form technical writing related to Oracle technologies, artificial intelligence, and natural language processing.\nNatural Language Processing on Oracle Cloud Infrastructure A practical book on building and scaling NLP solutions on Oracle Cloud Infrastructure, with a focus on real architectures and applied enterprise use cases.\nOracle AI and Database Engineering My more recent writing focuses on Oracle Database AI capabilities, including vector search, ONNX workflows, semantic retrieval, and implementation patterns for modern AI-enabled applications.\nAs new books, chapters, and technical contributions become available, I will list them here.\n","permalink":"https://assoudi.blog/books/","summary":"\u003cp\u003eThis page gathers my books and long-form technical writing related to Oracle technologies, artificial intelligence, and natural language processing.\u003c/p\u003e\n\u003ch2 id=\"natural-language-processing-on-oracle-cloud-infrastructure\"\u003eNatural Language Processing on Oracle Cloud Infrastructure\u003c/h2\u003e\n\u003cimg src=\"/images/books/nlp_on_oci.jpg\" alt=\"Natural Language Processing on Oracle Cloud Infrastructure\" width=\"220\"\u003e\n\u003cp\u003eA practical book on building and scaling NLP solutions on Oracle Cloud Infrastructure, with a focus on real architectures and applied enterprise use cases.\u003c/p\u003e\n\u003ch2 id=\"oracle-ai-and-database-engineering\"\u003eOracle AI and Database Engineering\u003c/h2\u003e\n\u003cimg src=\"/images/books/oracle_ai_guide.jpg\" alt=\"Data Engineer's Guide to Oracle Machine Learning and GenAI Services: Modern data engineering practices for creating efficient AI-driven applications at enterprise scale\" width=\"220\"\u003e\n\u003cp\u003eMy more recent writing focuses on Oracle Database AI capabilities, including vector search, ONNX workflows, semantic retrieval, and implementation patterns for modern AI-enabled applications.\u003c/p\u003e","title":"Books"},{"content":" A practical local setup for validating Oracle AI code, testing vector-search workflows, and building a repeatable 26ai experimentation environment.\nSetting up a local Oracle Database 26ai Free environment is one of the most practical ways to start experimenting seriously with the new AI-related capabilities of the database without waiting for access to a larger managed environment.\nWhile working on chapters for my new co-authored Packt book on Oracle AI, I needed a repeatable setup to validate code, test ideas, and refine examples before they made their way into the manuscript. A local Docker-based lab turned out to be the simplest and most effective answer. It gave me a controlled environment for testing AI Vector Search, ONNX model loading, SQL-based embeddings, and early database-side RAG experiments without introducing unnecessary operational overhead.\nThis article documents the setup I used on Windows. The goal is not to build a production deployment. The goal is to create a lightweight and reusable local lab that becomes part of your daily experimentation workflow.\nWhy this local lab matters A local Oracle Database 26ai Free lab is useful because it removes friction.\nInstead of depending on a remote environment every time you want to test a query, validate a notebook integration, or try a new vector-search pattern, you can work locally with a disposable but persistent setup. In my case, this was especially useful during writing and validation work, where I needed to repeatedly test snippets and small workflow variations.\nFor practitioners working through the new Oracle AI capabilities, this kind of setup is valuable for at least three reasons:\nit is fast to rebuild it is cheap to run it creates a stable sandbox for experimentation before moving to a more formal environment The setup I used The environment was intentionally simple:\nWindows as the host machine Docker to run Oracle Database 26ai Free locally SQL Developer for database-side work VS Code for Python and notebook-based integration where needed That combination is more than enough for a serious local lab.\nStep 1: Pull the Oracle Database 26ai Free image Start by pulling the Oracle Database Free image from Oracle Container Registry:\n1 docker pull container-registry.oracle.com/database/free:latest This gives you the local image you need to build the environment and serves as the base for the rest of the setup.\nStep 2: Start the container with persistence Once the image is available locally, start the container with the database ports exposed and a local folder mounted for persistence.\nOn Windows, I used this pattern:\n1 2 3 4 5 6 7 docker run -d ^ --name oracle-26ai-free ^ -p 1521:1521 ^ -p 5500:5500 ^ -e ORACLE_PWD=password123 ^ -v \u0026#34;%cd%:/opt/oracle/oradata\u0026#34; ^ container-registry.oracle.com/database/free:latest A few details matter here:\n1521 is the main database listener port 5500 can be useful for additional management access ORACLE_PWD sets the database password during initialization the volume mount to /opt/oracle/oradata preserves the database files across container restarts That persistence is not a minor detail. It is what makes the setup genuinely usable for ongoing work. Without it, each rebuild becomes much less attractive because you lose your database state and have to recreate the environment from scratch.\nStep 3: Verify that the database is healthy After starting the container, do not rush directly into client tools. First make sure the database is fully initialized.\nA quick first check is:\n1 docker ps That confirms the container is up, but for Oracle Database startup, that alone is not enough. It is also worth checking the logs either from Docker Desktop or directly from the command line:\n1 docker logs -f oracle-26ai-free In my setup, I used Docker Desktop logs as a quick visual confirmation that the container was not only running, but that the Oracle database inside it had fully initialized and was ready for connections.\nStep 4: Test connectivity from inside the container Before connecting from external tools, validate the database from inside the container with sqlplus:\n1 docker exec -it oracle-26ai-free sqlplus pdbadmin/password123@FREEPDB1 Once connected, a simple query such as:\n1 SELECT 1 FROM dual; is enough to confirm that the session is alive and the service is reachable. This is a small step, but it helps separate database startup issues from client-side connection issues.\nStep 5: Connect from SQL Developer Once the internal validation works, connect from SQL Developer using:\nUser: pdbadmin Password: password123 Host: localhost Port: 1521 Service name: FREEPDB1 This is the point where the setup stops feeling like “just a container” and starts becoming a normal working Oracle environment. From here, you can write SQL, test vector-related features, load models, and start building real local experiments.\nWhat this setup is good for This local lab is especially useful for:\nvalidating Oracle Database 26ai code snippets testing AI Vector Search workflows experimenting with ONNX model loading trying SQL-based embeddings preparing notebook or Python integration experiments building the base environment for later RAG-style database experiments In other words, it is not just a setup tutorial. It is the starting point for a whole series of practical 26ai experiments.\nA few practical notes Even in a lightweight local setup, a few habits are worth keeping:\nuse persistence from the beginning verify logs before assuming the database is ready test internal connectivity before debugging client tools keep the environment simple in the first iteration For a local lab, simplicity is a feature. The point is not to simulate every production concern at once. The point is to create a stable sandbox you will actually use.\nConclusion This Oracle Database 26ai Free setup on Docker proved to be a very practical local lab: lightweight, repeatable, and no-cost. It gave me a solid foundation for validating code, testing ideas, and working hands-on with Oracle’s evolving AI-related features in a controlled environment.\nWith that foundation in place, the next logical step is to move beyond setup and into actual experimentation. In the next article in this series, I will build on this lab to explore embeddings and semantic similarity search in Oracle Database 26ai.\n","permalink":"https://assoudi.blog/posts/building-local-oracle-database-26ai-free-lab/","summary":"\u003cblockquote\u003e\n\u003cp\u003e\u003cstrong\u003eA practical local setup for validating Oracle AI code, testing vector-search workflows, and building a repeatable 26ai experimentation environment.\u003c/strong\u003e\u003c/p\u003e\n\u003c/blockquote\u003e\n\u003cp\u003eSetting up a local Oracle Database 26ai Free environment is one of the most practical ways to start experimenting seriously with the new AI-related capabilities of the database without waiting for access to a larger managed environment.\u003c/p\u003e\n\u003cp\u003eWhile working on chapters for my new co-authored Packt book on Oracle AI, I needed a repeatable setup to validate code, test ideas, and refine examples before they made their way into the manuscript. A local Docker-based lab turned out to be the simplest and most effective answer. It gave me a controlled environment for testing AI Vector Search, ONNX model loading, SQL-based embeddings, and early database-side RAG experiments without introducing unnecessary operational overhead.\u003c/p\u003e","title":"Building a Local Oracle Database 26ai Free Lab with Docker on Windows"},{"content":"This page gathers selected technical resources related to my work on Oracle + AI.\nOver time, it will include:\nGitHub repositories Hugging Face models and datasets technical walkthroughs and videos companion resources for articles implementation templates and reference material External profiles GitHub Hugging Face LinkedIn Typica.ai ","permalink":"https://assoudi.blog/resources/","summary":"\u003cp\u003eThis page gathers selected technical resources related to my work on Oracle + AI.\u003c/p\u003e\n\u003cp\u003eOver time, it will include:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eGitHub repositories\u003c/li\u003e\n\u003cli\u003eHugging Face models and datasets\u003c/li\u003e\n\u003cli\u003etechnical walkthroughs and videos\u003c/li\u003e\n\u003cli\u003ecompanion resources for articles\u003c/li\u003e\n\u003cli\u003eimplementation templates and reference material\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"external-profiles\"\u003eExternal profiles\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003eGitHub\u003c/li\u003e\n\u003cli\u003eHugging Face\u003c/li\u003e\n\u003cli\u003eLinkedIn\u003c/li\u003e\n\u003cli\u003eTypica.ai\u003c/li\u003e\n\u003c/ul\u003e","title":"Resources"},{"content":"If you would like to receive future articles and updates on Oracle AI, vector search, ONNX workflows, and enterprise implementation patterns, a newsletter option will be available here soon.\nIn the meantime, you can follow my work through this blog and my public channels.\n","permalink":"https://assoudi.blog/subscribe/","summary":"\u003cp\u003eIf you would like to receive future articles and updates on Oracle AI, vector search, ONNX workflows, and enterprise implementation patterns, a newsletter option will be available here soon.\u003c/p\u003e\n\u003cp\u003eIn the meantime, you can follow my work through this blog and my public channels.\u003c/p\u003e","title":"Subscribe"},{"content":"Welcome to assoudi.blog.\nThis is where I publish practical, technically grounded writing on Oracle AI.\nMy goal is simple: useful Oracle + AI content for practitioners, architects, and technical leaders.\n","permalink":"https://assoudi.blog/posts/welcome-to-assoudi-blog/","summary":"\u003cp\u003eWelcome to assoudi.blog.\u003c/p\u003e\n\u003cp\u003eThis is where I publish practical, technically grounded writing on Oracle AI.\u003c/p\u003e\n\u003cp\u003eMy goal is simple: useful Oracle + AI content for practitioners, architects, and technical leaders.\u003c/p\u003e","title":"Welcome to assoudi.blog"}]