Victor Alencar — writing from the field
Field notes on building AI systems in production.
Deep dives on AI, cloud, and software engineering — SageMaker pipelines, Bedrock agents, embeddings, and the messy details in between.
Latest writing
7 entries-
007
The Fleet — How to Actually Use AI as a Software Engineer
A practical playbook for going from "AI as autocomplete" to running a fleet — work loops, workflows, parallel subagents, and multiple Claude Code sessions across tmux panes.
-
006
Building Company-Aware AI Agents on AWS Bedrock with AgentCore and MCP
How to build an agent on Bedrock AgentCore that connects to your company's tools over MCP, grounds itself in internal documents with a Knowledge Base, and ships as a quick Python chatbot.
-
005
Using LLM Embeddings to Transform Risk Data Analysis
How embeddings work under the hood and how to leverage them to build intelligent, user-friendly outputs from your trained models and risk datasets.
-
004
Building an End-to-End LLM Pipeline with AWS SageMaker
How to train and deploy a large language model using SageMaker, Glue, and Athena — from raw data extraction all the way to live inference endpoints.
-
003
Deploying to Netlify - A Beginner's Guide
Step-by-step guide to deploying your static site to Netlify for free.