Let’s talk

MIGUEL RODRIGUEZ / DATA, CODE & DESIGN

I turn messy datainto useful tools.

Senior data analyst. Automation builder.
A designer who can also build the tool.

Follow one invoice View résumé

Alteryx billing workflow~9 hours → 20 minutes/month

Alteryx · Python · Power BI · React / TypeScript
Reported 2023 result · initial setup separate

01 / BILLING AUTOMATION

Follow the record.
Find the difference.

DEMO-104 · From a vendor export to a reviewable $160 mismatch.

01 / Vendor export.xlsx
RecordVendorBilled
DEMO-102 Juniper LLC$1,080
DEMO-103ALDER-co$1,200
DEMO-104NORTH_CO $2,140
DEMO-105 Juniper LLC$1,440

Different names. Different date formats.

02 / Map the fields
NORTH_CO North CoJan-26 2026-01DEMO-104 stays with the row.

A consistent schema for the next file.

03 / Match the records
key = vendor + period
    + invoice_id

join(export, ledger, key)
delta = billed - expected

Simplified logic · source stays attached.

04 / The exception
DEMO-104
Billed$2,140
Expected$1,980
Difference+$160

Flag the difference for analyst review.

05 / Review coverage
Hover, focus or tap a point
0 checked12 checked

9 matches. 3 items that need a decision.

06 / Review queue
104 Amount mismatch107 Duplicate key109 Missing account

The three exceptions, with their sources.

LOCKTON · BILLING AUTOMATION

One monthly task.
Hours given back.

The exception is visible. The monthly rebuild is gone.

Monthly processing timeLower is better
≈96%

less monthly
processing time

THE HIGH-LEVEL IMPACT

8 hours 40 minutes back per monthly run.

Clean the inputs, consolidate invoices, surface discrepancies—and put the recovered time into client decisions.

Reported in my Alteryx Day presentation, 2023. One workflow example; initial setup is separate. All invoice cards use fictional sample data.

Scroll to follow the record
Skip story
01 / THE BILLING WORKFLOW

Find the exception.
Give the time back.

Every month, the same spreadsheet cleanup stood between an analyst and the real work. I built an Alteryx workflow that consolidated vendor data and put discrepancies in view.

MY PART

Workflow design, vendor and month controls, data preparation, reconciliation, and analyst enablement.

See the build and reported result
MONTH-END / JANUARY 2026

Follow the invoice.

Fictional sample
NORTH_CO Jan-26 · DEMO-104North Co2026-01 · DEMO-104+$160DEMO-104 · Needs review
12records imported
9matched
3need review
The same month arrives three ways. Keep the original row so every change can be traced.
InvoiceVendorSource periodBilledFile
NORTH_CO Jan-26$960vendor.xlsx
Juniper LLCJAN 2026$1,080export.csv
ALDER-co2026/01$1,200billing.xlsx
NORTH_CO Jan-26$2,140vendor.xlsx
Juniper LLCJAN 2026$1,440export.csv
ALDER-co2026/01$1,560billing.xlsx
NORTH_CO Jan-26$1,680vendor.xlsx
Juniper LLCJAN 2026$1,800export.csv
ALDER-co2026/01$1,920billing.xlsx
NORTH_CO Jan-26$2,040vendor.xlsx
NORTH_CO JAN 2026$1,680export.csv
ALDER-co2026/01$2,280billing.xlsx
DEMO-104 / Amount mismatch

The difference is the next action.

Billed$2,140Expected$1,980Difference+$160

The billed amount is $160 above the ledger. Check the vendor rate before approving the invoice.

See the transformation logic 
vendor = title_case(trim(replace_separators(raw.vendor)))
period = parse_month(raw.period)
key = (vendor, period, raw.invoice_id)

expected = ledger.lookup(key)
delta = raw.billed - expected.amount
review_if(duplicate(key) or delta != 0 or missing(account))

Original simplified example. Alteryx powered the reported workflow; Power Query is also part of my data-preparation practice.

12 fictional rows demonstrate the process. The reported ~9 hours → 20 minutes/month comes from my separate 2023 Alteryx workflow.

02 / EMPLOYEE SENTIMENT INTELLIGENCE

A rating is a number.
A comparison gives it meaning.

A score alone cannot tell a team where to focus. I connected research, source-specific parsing, Power BI comparisons, and reporting so the context traveled with the numbers.

MY PART

Python research tools, metric definitions, dashboard design, product offerings, and executive presentations.

Follow the ESA product story
ESA / RESEARCH RATINGS

A score with context.

Fictional companies
FOCUS COMPANYCedar Works
Source: research samplePeriod: Q1 2026Unit: rating, 1–5

How the sample rating changed

24 months · 1–5 scale
12345Apr 2024Apr 2025Mar 2026
Hover, focus or tap a point
01 / PRESERVE THE SOURCE
Cedar Works / Career Opportunities
Captured rating
3.2 / 5
Source reference
research-sample / cedar / Q1-2026
Missing distribution
Unavailable · not inferred from a mean
Select a dimension
12345
Cedar WorksSelected peer average
Career Opportunities

A gap worth investigating.

-0.60 points vs selected peers
FROM CHECKED VALUES TO AN EDITABLE BRIEF

Career Opportunities

Cedar Works3.2 / 5Selected peer average3.80 / 5Gap-0.60 pts
Values are calculated from the sample. The observation remains a draft for analyst review.
Inspect the ratings and calculation
CompanyCareer OpportunitiesIncluded
Cedar Works3.2 / 5Focus
Juniper3.9 / 5Selected peer
Alder3.7 / 5Selected peer
Maple3.8 / 5Selected peer
peer_average = mean(selected_peers.same_scale_ratings)
gap = company_rating - peer_average
report.chart_values = checked_values
report.observation = draft_for_analyst_review

The peer average is an unweighted mean of the selected company ratings. It is not an industry benchmark.

Original illustrative reconstruction of the earlier research-rating workflow. The newer Voice Lab reporting app uses a different percentage scale and remains in development.

03 / AI ORCHESTRATION

Follow the work
behind the answer.

More agents can mean more unsupported claims. I build systems where research, critique and review have different jobs—and every finding can be traced back to a source.

MY PART

Product interfaces, Python workflow contracts, role-based orchestration, evidence ledgers, and review states.

Explore TradeLab research Explore the local AI platform
RESEARCH / ILLUSTRATIVE REPLAY

Follow the claim.

Human review
“Did the new guide improve onboarding?”
FROM SOURCE TO SOMETHING USEFULExplore a layer
One connected system.

Data + code + design

Checked values

Validate the calculation before it becomes a claim. A useful answer separates what is known from what still needs review.

WORKING HYPOTHESIS

The guide improved onboarding.

The initial claim is plausible, but it needs testing against the source material.

CHECK THIS BEFORE CONCLUDINGCompletion time before and after launch.
E-01 / SOURCE DETAIL

The release adds an onboarding guide.

06 May · Fictional research evidence
See how a finding keeps its source 
{
  "claim": "Guide adoption in sample: 60%",
  "evidence": ["E-02"],
  "calculation": "18 / 30",
  "review": "supported",
  "unknown": "effect on completion time"
}

A fictional research example informed by TradeLab and local orchestration work. It runs entirely in this page.