latentExplorations in latent space
Writing about machine learning, generative AI, and high-performance engineering. Principal DBA & MCS–DS graduate from UIUC.
Articles
Deep dives into data science, machine learning, and AI engineering.
Running Ollama Right: Local Models and Your Own Fine-Tunes on One GPU
Ollama is the least painful way to serve local models on a single GPU, but the defaults are tuned for demos, not for work. What actually matters — quantization, context memory, keep-alive — and how to get a fine-tuned model of your own running behind the same API.
I Put an AI in My World Cup Pool. Here's How It Actually Did.
A bot that reads live match status off the open web, turns a qualitative read of each game into a scoreline, and gets graded against real results every day. The group stage is over, all 72 matches are graded, and the bot is on top of the pool.
Two Pi Agents and a Customer Table: Masking PII Before It Crosses the Wire
When one agent owns the sensitive data and another needs a copy, the safe move is to mask at the producer, not to trust the consumer. I built it with two Pi agents and a bank dataset, then grepped the wire for leaks. The count came back zero.
How a ReAct Agent Loop Actually Works
A walkthrough of one real ReAct agent run — the query was "Houston, we have a problem" and the agent returned the exact YouTube timestamp where Tom Hanks delivers the line. You can replay the recorded run right inside this article, thinking tokens and all.
Prompt Engineering is Dead, Long Live Prompt Engineering
Everyone said prompt engineering was a fad. They were wrong — it just evolved. From artisanal prompting to systematic prompt design for production systems.
Building AI-Powered Personal Websites
Your portfolio site can do more than display static content. Learn how to integrate AI chat, RAG, and agentic tools into a personal website.
The Rise of Structured Generation: From JSON Mode to Grammar-Constrained Decoding
Guaranteeing valid output from LLMs requires more than prompting. Grammar-constrained decoding enforces structure at the token level — here's how it works.
Self-Hosting AI: Running LLMs on Your Own Hardware
Cloud APIs are convenient but expensive. Explore how to run open-source LLMs on your own servers — from hardware selection to inference optimization.
PostgreSQL as a Vector Database: pgvector in Production
You don't need a separate vector database. pgvector turns PostgreSQL into a semantic search engine — with HNSW indexes, hybrid queries, and full SQL power.
Claude Code and the Future of AI-Assisted Development
Claude Code brings an AI agent directly into your terminal. Explore what autonomous coding tools mean for software engineering workflows.
Agentic Workflows: Orchestrating Multi-Step AI Pipelines
Single-prompt AI is hitting its ceiling. Agentic workflows chain multiple LLM calls with tools, branching, and feedback loops to tackle complex tasks reliably.
MCP Servers: Building Tool-Using AI with Model Context Protocol
The Model Context Protocol standardizes how AI models discover and use tools. Here's how MCP servers work and why they matter for the agentic future.
12 of 41 articles
Ask my AI
Trained on my resume, GitHub projects, and blog articles. Ask about my background, paste a job description for fit analysis, or drop a database error log to see the expert system in action.
AI Assistant
Powered by Gemini · RAG · agentic tools
Hi! I'm an AI assistant for this blog.
I can answer questions about Oswaldo, but I can also take action for you right here:
- Schedule a call — I'll check his calendar and send you a Google Calendar invite
- Send a message — share your contact info and I'll email it to him directly
- Generate a tailored resume — tell me the role and I'll produce a targeted version
What would you like to know or do?
About
I'm Oswaldo Orona — a Principal Database Administrator and AI practitioner based in Denver, CO. I hold an MCS–DS from UIUC (Tau Beta Pi, Phi Kappa Phi) and bring 25+ years of database experience alongside deep hands-on work in machine learning and AI engineering.
My focus areas include Retrieval-Augmented Generation (RAG), AI agents, Model Context Protocol (MCP), geospatial AI, and financial AI — all running in a self-hosted Proxmox home lab with Docker and LXC.
This blog is where I document explorations in latent space: the ideas, experiments, and systems that live between the data and the model.
ML & AI
- PyTorch
- TensorFlow
- RAG
- NLP
- Computer Vision
- LLMs
Databases
- PostgreSQL
- Oracle
- Redis
- pgvector
- DynamoDB
Infrastructure
- Docker
- Proxmox
- Ansible
- AWS
- Linux
Languages
- Python
- R
- SQL
- PL/SQL
- Java
- Bash