---
title: "Harvey Ducay (@haharvs) | Hashnode"
description: "Aiming to develop advanced data systems for AI and everything in between. Read the latest articles by Harvey Ducay on Hashnode."
url: "https://hashnode.com/@haharvs"
---

@haharvs

# Harvey Ducay

@haharvs · Manila, Philippines · Joined September 2024

Aiming to develop advanced data systems for AI and everything in between.

[Website](https://harveyducay.dev/)

### [The Two Numbers That Predict AI Agent Reliability](https://hddatascience.tech/the-two-numbers-that-predict-ai-agent-reliability?utm_source=hashnode&utm_medium=feed)

Back then, I felt lost while writing automated Playwright tests. I'd turned to Claude Code to help speed up the process, but I kept hitting the same wall. The generated tests were redundant, surface-l

10 min read · [hddatascience.tech](https://hddatascience.tech/the-two-numbers-that-predict-ai-agent-reliability?utm_source=hashnode&utm_medium=feed)

[Apr 25](https://hashnode.com/posts/the-two-numbers-that-predict-ai-agent-reliability/69eca78d17d72fa4cd7e75a8)

### [The LLM Council and the Human Mind](https://hddatascience.tech/the-llm-council-and-the-human-mind?utm_source=hashnode&utm_medium=feed)

Understanding prayer and meditation as a multi-agent debate.

5 min read · [hddatascience.tech](https://hddatascience.tech/the-llm-council-and-the-human-mind?utm_source=hashnode&utm_medium=feed)

[Dec 16, 2025](https://hashnode.com/posts/the-llm-council-and-the-human-mind/6940ab3bebb635d83efb5989)

### [One-Shot Trauma: When Reinforcement Learning and Human Minds Overcorrect](https://hddatascience.tech/one-shot-trauma-when-reinforcement-learning-and-human-minds-overcorrect?utm_source=hashnode&utm_medium=feed)

My experience with choking taught me more about the limits of AI than any textbook could.

7 min read · [hddatascience.tech](https://hddatascience.tech/one-shot-trauma-when-reinforcement-learning-and-human-minds-overcorrect?utm_source=hashnode&utm_medium=feed)

[Nov 17, 2025](https://hashnode.com/posts/one-shot-trauma-when-reinforcement-learning-and-human-minds-overcorrect/691b2b472cd7052af34c952e)

### [What is AI?](https://hddatascience.tech/what-is-ai?utm_source=hashnode&utm_medium=feed)

An Applied Mathematician's Attempt at a Rigorous Answer.

4 min read · [hddatascience.tech](https://hddatascience.tech/what-is-ai?utm_source=hashnode&utm_medium=feed)

[Nov 2, 2025](https://hashnode.com/posts/what-is-ai/690711076e91cce3524d0396)

### [From "It Works" to "Why It Works": A Call for Deeper Understanding in Data Science](https://hddatascience.tech/from-it-works-to-why-it-works-a-call-for-deeper-understanding-in-data-science?utm_source=hashnode&utm_medium=feed)

A Deep Dive in Using Convolutional Neural Networks

4 min read · [hddatascience.tech](https://hddatascience.tech/from-it-works-to-why-it-works-a-call-for-deeper-understanding-in-data-science?utm_source=hashnode&utm_medium=feed)

[Oct 5, 2025](https://hashnode.com/posts/from-it-works-to-why-it-works-a-call-for-deeper-understanding-in-data-science/68e209ff4394a8d6c4f3b620)