PrachiVerma

Hello, I’m Prachi.

I like figuring out how things work—and finding ways to make them better.

I get curious, learn what I need, and tend to make something useful from the answer. Sometimes that means building a product. Sometimes it means finding a clearer way for people to work.

Today, most of my professional work is in analytics leadership.

View selected work

Selected work

Three ways curiosity became concrete.

A professional system, an independent product, and a venture. Different forms of work; the same instinct to make something useful.

Professional system · Public-safe reconstruction

Making complex analytics work easier to run

Flow: requests are structured through explicit ownership, priority, and context, then move through quality checks to delivery. A generalized view of the operating logic. Illustrative reconstruction; no client data.
The problem
Work spanning multiple markets had to move without losing the context held by individual analysts: who owned what, what mattered most, and where quality could fail.
What I built
I designed a clearer operating system for inventory, ownership, handover, quality checks, and escalation. The useful part was not another dashboard. It was making the work easier to understand and run.
My role
I led the operating design and quality approach. Analysts and partner teams executed the underlying delivery work.
Where it stands
The system was designed to make ownership, handover, and quality expectations explicit. Exact operational details remain confidential.

Read the public-safe case

Independent Product · In Development

AI-assisted Rental Condition Report

Move-in and move-out condition records are often inconsistent, hard to compare, and too easy to dispute later.

  1. 01CaptureRoom, item, and photo
  2. 02SuggestAI-assisted observation
  3. 03ReviewA person accepts or edits
  4. 04RecordOnly reviewed text is saved
Proposed inspection flow. Product implementation is in progress.
What I built
I defined the product requirements, minimum viable scope, inspection workflow, and data model. AI can suggest observations from photos, but a person must review them before they become part of the record.
My role
I am shaping the product and learning the technical pieces needed to build it.
Current status
Parts of the product are being implemented. It is not released, and there are no user, adoption, or revenue claims yet.

Read the product case

Venture · Co-Founder

Gâteaux d’Amour

Turning a product idea into a real consumer business meant understanding what customers wanted—and then making the whole operation work around it.

3 years, 2 months
operating the ventureFebruary 2019–March 2022
20+
offeringsshaped by feedback and demand

We used customer feedback, market research, and demand patterns to shape the range. I worked across product strategy, customer experience, marketing, and day-to-day operations.

I co-founded and operated the venture.

We ran the business from February 2019 to March 2022 and developed more than 20 offerings.

Read the Gâteaux story

What I do today

Analytics is my deepest professional specialization right now.

I lead analytical work across complex, multi-market environments. That includes deciding what the real question is, creating structure around delivery, keeping quality visible, and helping teams do strong work without losing the context behind it.

The tools matter. Judgment about what to build, how to run it, and what not to claim matters more.

More systems

Job Search OS

My own job search had too many moving parts, so I started turning one repetitive part into a system.

The wider system is still being documented. This prototype shows only the component that can be verified today.

Job Search OS
Working subsystem · Job Listing AgentCollect → filter → organize
Several listing sourcesPersistent tracker
The implemented agent collects listings from several sources, normalizes and filters them, removes duplicates, classifies relevant roles, and organizes the results in a persistent tracker.

How I tend to work

Less of a framework. More of a recurring habit.

  1. Start with the real question

    I keep asking until I understand the problem behind the original request.

  2. Make the answer tangible

    I learn what I need and build the useful thing—a product, workflow, business, or clearer way to work.

  3. Keep judgment visible

    I am specific about my role, the team’s work, and where a person still needs to decide.

  4. Notice what happens next

    I look at what changed, what did not, and what I would do differently the next time.

Behind the work

The work changes. The way I approach it keeps showing up.

Different chapters have asked me to learn different things: how to run a business, lead analytical work, design better workflows, and build software. I am interested in the moment when a question becomes concrete enough to make—and useful enough to keep improving.

More about how I work