@haharvs

Harvey Ducay

@haharvs

Website

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

Manila, PhilippinesJoined September 2024

About

Nothing here yet.

Available for

Nothing here yet.

Harvey Ducay's blogs

Data Science/Engineering Insightshddatascience.tech25 posts

The v0.dev Hubharveyducay.blog15 posts

Articles Comments

Recently published

HDHarvey Ducayhddatascience.techApr 25 · 10 min read

The Two Numbers That Predict AI Agent Reliability

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

HDHarvey Ducayhddatascience.techDec 16, 2025 · 5 min read

The LLM Council and the Human Mind

Weeks ago, Andrej Karpathy(ex. director of AI at tesla) launched LLM Council. The concept is brilliant but simple: instead of asking a single AI model (like ChatGPT) to answer a question, you create a "council" of different models. You have one model...

HDHarvey Ducayhddatascience.techNov 17, 2025 · 7 min read

One-Shot Trauma: When Reinforcement Learning and Human Minds Overcorrect

The Day My Internal Agent Received a -1,000,000 Penalty It only took a second to rewire my brain. By early 2022, I was just your average joe, living life day by day. Eating was one of my daily tasks that was necessary, automatic, and unconscious. It ...

HDHarvey Ducayhddatascience.techNov 2, 2025 · 4 min read

What is AI?

I find it a little funny that I'm only getting to this article now. As an applied mathematician, I have a deep-seated need for rigor, for building arguments from first principles. In a field as dynamic and hype-driven as Artificial Intelligence, a ri...

HDHarvey Ducayhddatascience.techOct 5, 2025 · 4 min read

From "It Works" to "Why It Works": A Call for Deeper Understanding in Data Science

Sometimes, the most valuable lessons come from unexpected moments. I was attending a data science workshop recently, and a brief discussion served as a powerful reminder of a crucial question we must ask ourselves: are we content with knowing that so...

Harvey Ducay

About

Available for

Harvey Ducay's blogs

Recently published

The Two Numbers That Predict AI Agent Reliability

The LLM Council and the Human Mind

One-Shot Trauma: When Reinforcement Learning and Human Minds Overcorrect

What is AI?

From "It Works" to "Why It Works": A Call for Deeper Understanding in Data Science

Search Hashnode

Harvey Ducay

About

Available for

Harvey Ducay's blogs

Recently published

The Two Numbers That Predict AI Agent Reliability

The LLM Council and the Human Mind

One-Shot Trauma: When Reinforcement Learning and Human Minds Overcorrect

What is AI?

From "It Works" to "Why It Works": A Call for Deeper Understanding in Data Science