Skip to content
View wanwrick's full-sized avatar
0 followers · 1 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report wanwrick

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
wanwrick/README.md

Paroz Mehta

Manager, Forward Deployed Engineering (Producers) at Questrade Financial Group Toronto, Ontario

What I work on

I lead a client-facing data engineering squad inside a financial group. We take a business priority, ship the data product that answers it, then hand the pattern back to the platform team so the next team does not need us.

The work spans a regulated digital banking launch, a lakehouse migration across two clouds, and the governance layer that has to hold up to an auditor. Most of what I do now is deciding which problems are worth a platform and which are worth a pattern.

Selected work

Repository The question it answers
platform-investment-case Build the lakehouse, extend the warehouse, or buy the vendor product? A DCF, a decision tree and a Monte Carlo find that the decision turns on adoption, not architecture.
pipeline-sla-capacity How much capacity holds a 15-minute freshness SLA? Queueing theory says none does, because the service times alone breach it.
medallion-pipeline How do raw events become governed, business-ready tables without a hand-built pipeline per source, and how do you know when they stop being either?
ai-data-analyst Can a business user get a trustworthy answer from the warehouse in plain language, with the SQL and permissions shown?
noosphere-public What does an AI work partner need to know about your business before it is useful?

The first two answer a money question and an operations question. The rest are the platform underneath them. Each runs on synthetic data and carries no employer detail.

Background

Dual Executive MBA, Cornell Johnson and Queen's Smith, conferred May 2026. Databricks, PMI, Scrum Alliance, and Google Cloud credentials.

Platform: Databricks on Azure and GCP, Delta Lake, Unity Catalog, BigQuery, Pub/Sub, Kafka, Power BI, Python, SQL.

Contact

LinkedIn · parozmehta@gmail.com · Toronto, Ontario

Pinned Loading

  1. ai-data-analyst ai-data-analyst Public

    Ask your warehouse a question in plain language. Every answer comes back with the SQL that produced it, its sources, and the permissions it ran under. Runs in demo mode with no credentials.

    Python

  2. medallion-pipeline medallion-pipeline Public

    Production-grade Medallion Architecture (Bronze/Silver/Gold) on Databricks with DLT, CDC, SCD Type 2, Unity Catalog governance, and Asset Bundles

    Python 1

  3. noosphere-public noosphere-public Public template

    The context layer for your AI work partner. 32 files, four layers, zero dependencies. Fork it, customize it, make it yours.

    Python

  4. platform-investment-case platform-investment-case Public

    Build, extend, or buy? A DCF, decision tree and Monte Carlo over the same scenarios, generating the recommendation memo. The decision turns on adoption, not architecture.

    Python

  5. pipeline-sla-capacity pipeline-sla-capacity Public

    How much capacity holds a 15-minute freshness SLA? Little's Law, Kingman's VUT and discrete-event simulation say: none, and here is why.

    Python