Databricks SQL & data engineering
Layered CTEs, multi-source joins, customer primary-key unification, and verification against the live implementation. The bugs get found before the campaign does.
Rtvik Nambiar · Toronto · Canadian citizen
I came up building data pipelines. Now I run the releases that ship them.
Technical Lead at HCLTech, running program management on a customer data platform migration for a Canadian credit card issuer. SAS CI 360 and Salesforce Marketing Cloud are going out. Databricks and Braze are coming in, across five releases. I still write the SQL.
3 of 5
releases under my program management on the current migration
15%
less pipeline downtime from SQL and Spark tuning · data engineering, 2021–24
30%
faster release cycles from CI/CD automation · same role
2.5 h → 30 min
per-hire clinical onboarding, 100+ hires · BI role, 2024–25
Capabilities
What five years covers.
Pipelines and analysis on one side, requirements and releases on the other. The chain is the point.
Program & release delivery
ELT approvals and go/no-go gating criteria for the final three releases of a five-release migration. Deployment days run on a briefed-stakeholder rhythm, not on hope.
Business rules into data logic
Abstract segmentation rules become queries an engineer can build and an auditor can defend. Documented across 20+ acquisition use cases.
Governance & PII
Access classification and source-to-target lineage for Customer 360 data, written so an audit finds answers instead of gaps.
Stakeholder alignment
Requirements gathered wide enough that the teams downstream of the platform stayed supported through a partial migration. Works in English and French.
AI tooling
Microsoft Copilot and Power Automate wired into daily delivery work, saving roughly 40% of documentation and reporting time. Tools change; the habit of automating the boring parts doesn't.
Selected work
One migration, five releases.
A Canadian credit card issuer is replacing SAS CI 360 and Salesforce Marketing Cloud with Databricks and Braze, on a medallion architecture. I started in the data. I now run program management for the final three releases.
PHASE 1
Acquisition-campaign migration Shipped
Acquisition campaigns moved first, setting the ingestion patterns every later phase reuses.
PHASE 2
Enhancements Shipped
Pipelines hardened and edge cases closed before the higher-stakes ingestion began.
PHASE 3
Hotfix & customer unification Shipped
Customer-unification ingestion through the medallion layers. The release went out clean: stakeholders briefed at every milestone, in-flight systems supported through a partial migration.
PHASE 4
Base-customer migration In delivery
The full base-customer population moves over. I run the ELT approvals and hold the gating criteria.
PHASE 5
Legacy decommission Ahead
The old stack retires and the platform becomes the single source of truth.
2.5 h → 30 min
Paper onboarding per clinical hire rebuilt in Power Apps, with progress saved on or off call. 100+ hires through it.
BI analyst · 2024–25 · EMS provider
15%
Less pipeline processing downtime from SQL and Spark tuning.
Data engineering · 2021–24
30%
Faster release cycles from Azure DevOps and Jenkins CI/CD.
Data engineering · 2021–24
~40%
Documentation and reporting time saved with Copilot and Power Automate.
Current program · 2025–26
Also built
Trident Rose
A quote follow-up tool for trades businesses running on Jobber. It watches every open quote and texts the ones that go quiet until the customer answers. Anything ambiguous goes to a human, never a guess. No clients yet. I built it to find out whether I could take a real problem and ship something small that solves it.
tridentrose.com ↗
A product, not a practice. For roles, use the email below.
About
From the pipelines to the program calendar.
I started as a data engineer, and that layer still anchors everything. When a release decision lands on my desk, I can open the notebook and check the logic myself. Business analysis taught me the other half of the job, which is that the hardest part of data work is rarely the data. It is agreeing on what the rules mean.
In between, I built BI and automation for emergency medical services, where shipping to non-technical users is the whole job. I studied Cognitive Systems at UBC, so I default to asking how a person will actually use the thing. English and French at work, Malayalam at home. I grew up across four countries, which mostly taught me to read a new room fast.
Where I'd like this to end up is the commercial and fan-data side of a football club. Clubs sit on some of the most fragmented customer data anywhere. I'm not there yet and I know what I still have to show.
Progression
Data Analyst Intern
UBC Sauder · 2021Data Engineer / Software Developer
2021–24 · pipelines, tuning, CI/CDBusiness Intelligence Analyst
Huly · 2024–25 · dashboards and automation for EMSData Platform Analyst, Governance & Change
2025 · same migration, requirements sideProgram Manager
now · closing the final three releases
BA Cognitive Systems, UBC
Toronto, ON
Canadian citizen
EN · FR bilingual