Tri_

Portfolio · Data, analytics, AI

From raw data to decisions.

Data, analytics, pipelines, and AI, built to be trusted. I'm Tri Juhari, and this is how I think about it.

How I work

  1. Step 01

    Understand

    Start from the decision someone has to make, then work backwards to the data it needs.

  2. Step 02

    Build

    Pipelines and models that are simple enough to explain and boring enough to trust.

  3. Step 03

    Check

    Test the numbers before anyone sees them. Wrong data fast is worse than right data slow.

  4. Step 04

    Explain

    Write down what was built and why, so the next person can pick it up.

What I build

Why did the weekly number drop?
Before: Three spreadsheets, three answers.
AfterOne definition of the metric, one source table, and a chart that shows the drop by segment.
The nightly load failed again.
Before: Manual reruns every morning.
AfterRetries, alerts, and a data check that stops bad rows before they reach reports.
Can we ask questions over our own docs?
Before: Search that returns the wrong page.
AfterAnswers that quote the source and say "no info" when nothing matches. Like the chat on this page.

Selected projects

All projects →
Ask my AI

Ask the portfolio anything.

It searches my CV and this site in your browser. Every answer is a real excerpt with a link to its source, and nothing is invented.

Ask about my projects, skills, or experience. I answer with excerpts from my CV and this site.

Quick questions

Where can I read your work?

The Projects page has short case studies, and the Blog has longer write-ups.

Is the AI chat a real LLM?

No. It searches my CV and this site in your browser and quotes what it finds, with a link to the source. If nothing matches, it says so.

How do I get in touch?

Use the contact form, or find me on GitHub, LinkedIn, or Medium via the icons below.

Do you have a CV?

Yes, there is a PDF on the Resume page.

contact.sh
$ send_message --to tri_juhari
Say hi →