Work
Operations and automation specialist
Business systems, reporting and workflow automation.
I find where the work actually breaks, then build the automation and the written process so it stays fixed. Most recently I was on the operations team for Mosaic's retail field-sales program for Google's Pixel NAMER division, working with four teams across the US, Canada and Mexico. Bilingual in English and Spanish.
- 8
- Production automations at Mosaic
- 250+
- Hours a year returned to the team
- 46% → 62%
- Fulfillment profitability in three months
Case studies
Problem, root cause, build, outcome
The full trace behind each number.
Fulfillment Turnaround
A cross-border fulfillment operation bleeding margin at the seams, traced to the handoffs, repaired in three months
- Customs-clearance delays
- Routine → ~90% eliminated
- Time to turnaround
- 3 months
Field-Rep Onboarding
Onboarding for 125+ field reps across three countries was slow. The delay lived in the sync between systems.
- Field reps covered
- 125+
- Systems reconciled
- 4, by hand → 1 automation
Executive Reporting
A weekly leadership deck built by hand from 6+ sources, and the reporting errors that came with it. The errors traced back to the handoffs.
- Critical reporting errors
- ~90% fewer
- Sources reconciled
- 6+
Strategy Exercise: SaaS Support Analysis
Root cause analysis of a SaaS support operation revealing a systemic failure chain, from sales pitch to customer crisis
- Reported problem
- Performance complaints
- Actual root cause
- Overbooking failure chain
At Mosaic
Eight production automations
Built in Google Apps Script, Python and BigQuery. Each one writes raw data only and leaves the math in visible formulas, so a non-technical reviewer can audit any number by clicking the cell. Every one shipped with an SOP, and I trained the team on them before I left.
- 01
Mexico promoter-deck pipeline
~90% less slide workAbout 10 Apps Script modules, built from zero in one week, that cut manual slide maintenance about 90% for the 172-store Red Zone pilot and its BigQuery-fed 5-slide executive deck.
- 02
Sales and inventory automation suite
2+ hrs a weekAggregation, daily channel sales and validation jobs that replaced 2+ hours of weekly manual consolidation, built so the team could own and run it.
- 03
Master Sellout ingest
Turns the weekly sell-out report into per-store line items, with merge logic that never lets an out-of-order ingest overwrite prior data.
- 04
Expense reconciliation (US + Canada)
30 min a weekPulls weekly carrier-card transactions, enriches each row against the field roster, and runs a row-level approval workflow. It removed about 30 minutes a week of manual work along with the whole class of reconciliation errors.
- 05
Red Zone Tracker (Python)
172 storesJoins the master store list, field visits, weekly sell-out and the field schedule to track 6 KPIs per retail chain across 172 stores.
- 06
Field-rep lifecycle and access
~60% fasterOnboarding, leave, return and offboarding for 125+ reps in three countries, cutting onboarding time about 60%, plus reconciliation of 200+ distribution lists.
- 07
Weekly executive reporting
Under 5 minPulls from 6+ sources and, in under 5 minutes, generates the 11-slide leadership deck that used to take 3+ hours to build by hand.
- 08
BigQuery-to-Sheets live pipeline
Scheduled incremental refreshes, automatic retries and anomaly alerts, so field managers work from current data instead of day-old snapshots.
Earlier builds
More automation work
Dealer Data Transformation Pipeline →
Mixed-format dealer data, validated and normalized automatically before anyone analyzes it.
User Groups Management System →
Managers make their own group changes through a self-service sheet, with an audit trail.
Quarterly Summary Generator →
The quarterly executive summary, built, formatted and sent automatically.
Independent builds
Products I built and run on my own time.
AI Cloud Agent System →
Autonomous agents handling interview coordination, reader engagement, scheduling, and live site chat, running a defense-in-depth guardrail stack
Celestium: Astrology Engine and App →
A native Swift app for iPhone, iPad and Mac, live on the App Store, built on a custom four-module astrology engine: natal computation, transit synthesis, astrocartography, human design and tarot. A With Jhinna company.
Background
Where the root-cause habit started →
A decade of bodywork and health coaching. The pain a client pointed to was often not where the problem started.