---
title: "About · Nick Tran"
url: https://trannolis.com/about
description: "Nick Tran, a data and platform engineer based in New York City. NYC-based data and platform engineer building and operating data, AI, and quantitative systems."
---

# About

I’m Nick, a data and platform engineer based in New York City. I build and operate data infrastructure, from cloud architecture and distributed systems to ML/AI platforms and quantitative research tools.

I studied computer science at NYU Tandon, with a minor in finance. That combination has stayed with me: I’m interested in how data becomes useful, and in the engineering that makes it dependable.

My work so far spans three domains, each with different data, controls and failure modes. Building those systems broadened my focus from individual pipelines to the platforms around them: how they’re deployed, how they fail, and how people use them in production.

-   **Mortgage lender** · Public · 2022: market data streaming for an MBS trading desk. [Case study](/work/market-data-streaming).
-   **Insurance broker** · Public · 2022–2026: the AWS and Databricks data lake, the first production ML platform, and tools for accessing governed data through AI.[Data lake](/work/data-lake-buildout), [AI over governed data](/work/ai-over-governed-data).
-   **Wealth management** · Startup · 2026 to present: data engineering.

From 2025 to mid-2026 I built [Pine Peak Associates](/work/pine-peak), an independent engineering and quantitative research practice: financial data infrastructure, research workflows, and AI-assisted tools, with reproducible research and audit trails. No capital-affecting or production action happened without human approval; research and paper trading ran with bounded autonomy inside evidence gates. I wound it down to join a wealth management startup, where I work on data today.

This is still a personal site: a place for technical notes, projects, and things I’m figuring out. The older posts are part of that history, so I’ve kept them in the [archive](/writing#archive).

## Ask me about

-   Governed data access for AI agents: MCP servers, semantic layers, per-user permissions.
-   Enforcing agent authority on the server instead of in the prompt.
-   Research gates for trading strategies: deflated Sharpe, out-of-sample seals, and how gate evidence breaks.
-   Streaming and lakehouse data platforms on AWS and Databricks.

## Facts

Based

New York City

Focus

Data platforms · AI agent infrastructure

Built

Pine Peak Associates (2025–2026)

Writing

Data systems and AI agents

Education

B.S. Computer Science, minor in Finance, NYU Tandon (2021)

Email

[nick.tran@pinepeakassociates.com](mailto:nick.tran@pinepeakassociates.com)

Elsewhere

[GitHub](https://github.com/trannolis) · [LinkedIn](https://www.linkedin.com/in/nicktran-1/)