Jerry R. Napier

Data enablement · Information governance · Analytics & automation

I turn complex systems into trusted ones.

I help teams turn business needs into reliable data, clear reporting, and tools that make everyday work easier. My background spans payments, consulting, and digital operations, with a focus on the people who use what I build.

View selected work View résumé (PDF)

Manager, Data Enablement at Publicis Health Media

Professional portrait of Jerry R. Napier
Jerry R. Napier Data Enablement & Information Governance
10+years across information & operations
Built for confident change
See an example

01 · From data to decisions

From fragmented project data to one decision-ready view.

To show how scattered project data can support clearer decisions, I built an account-analysis view using an anonymized synthetic test sample: 54 projects and $60.2M in modeled spend. The model highlights $20.2M in potential savings for further review—not realized savings.

54

sample projects organized into one account view

$60.2M

modeled spend structured for review

$20.2M

potential savings identified by the model, not realized savings

Anonymized synthetic test sampleDocumented business rulesRow-level traceabilityExecutive-ready review

Review the full account-analysis case study

Selected work

Built to be useful.

The interactive samples need JavaScript. You can still read the case notes and explore every project below.

Data checks, reporting tools, and workflows built around real working needs. Explore the samples and case notes to see what each project does and the decisions behind it.

DATA CONTRACT / 01 Public alpha · 0.3.4
01

Data quality · Python · CI

Data Contract Monitor

Public alpha · Source and release

A data-checking tool for teams that need reliable reports. It checks rules across files and compares runs, so a change in the data is easier to spot and explain. The same checks work locally and in automated workflows.

  • Exact cross-file checks
  • Run-to-run comparison
  • One shared rule engine
Open case note
Need
Catch quiet data failures before they become reporting, modeling, or operational problems.
What I built
A local-first validator for CSV, Excel, JSON, JSON Lines, and optional Parquet. One rule engine serves the dashboard, command line, API, package, and GitHub Action; streaming keeps supported checks exact across batches.
In practice
Compare runs and contract versions, trace findings to readable rules, and use the same results in local review and CI. A failed data contract stays distinct from a program failure.
Look inside
Sample YAML contracts and HTML, JSON, JUnit, and SARIF reports show how readable rules become findings a team can review. The source explains the exact checks used across files and batches.

Public alpha v0.3.4 · Synthetic demos · Trusted-workstation dashboard · Privacy signals require human review · Native Windows and endpoint-protection qualification remain separate release checks.

Explore Data Contract Monitor
BOTOPS / BROWSER DEMO Ready

Systems overviewReady for review

Simulated review · no access to your computer

Launcher auditReady

Health checksReady

Support exportReady

Browser-only sample

Python 3.10+ · Standard library · Windows

Ready to review three simulated operational signals.

02

Operations engineering · Python

BotOps Manager

Public source · Simulated browser sample

A Windows console for keeping track of independent automations. It brings status, startup checks, and troubleshooting details into one place. Before an approved action runs, it checks the project and process again to help avoid acting on the wrong one.

  • Project-scoped control
  • Identity checked at action
  • Redacted support exports
Open case note
Need
Make separate local automations easier to oversee while keeping each project’s controls clearly defined.
What I built
A Python console that brings launcher reviews, health signals, verified project controls, and redacted support reports together.
In practice
Operators can check status and troubleshooting details in one place. BotOps rechecks project identity before an approved action.
Look inside
Sample projects and simulated process records show how launcher reviews, identity checks, and support exports work before they are used with a real automation.

Python 3.10+ · Windows · Standard library only · No cloud service or installer required.

Explore BotOps Manager
CHI_ANALYSIS / 0313,333 records
Recorded failure share2010–2018
Highest Risk 1 share among classified60607 · 82.70%
03

Data analysis · Reproducibility

Chicago Food Inspection Outcomes

Reproducible analysis · 2010–2018 inspections

An analysis of 13,333 public inspection records that makes an easy-to-miss distinction: a facility’s risk category is not its inspection result. Separating the two helps readers interpret the findings, and the published workflow lets them follow the analysis themselves.

  • Explicit denominators
  • Reproducible analysis
  • Documented sources
Open case note
Need
Compare inspection outcomes and facility risk across three Chicago ZIP codes while keeping conclusions aligned with the study’s descriptive scope.
What I built
A bounded official-data snapshot, explicit study parameters, validation checks, a reproducible notebook, and a documented refresh path.
In practice
The analysis shows why outcome and facility-risk measures can lead to different follow-up questions.
Look inside
The notebook separates two findings: 60622 has the highest recorded failure share at 21.54%; 60607 has the highest Risk 1 share at 82.70% among classified records. The source snapshot and validation checks make the denominators traceable.

2010–2018 study period; source snapshot retrieved in July 2026. Records represent inspections, not unique restaurants. Risk 1 is a facility classification, not a failed inspection. This study does not rank restaurant safety.

Review the analysis
operations-review.log
04

Operations intelligence · Analytics

Operations Intelligence & Automation Platform

Public source · Synthetic demonstration

A service-operations demo for managers and analysts who need to move from a dashboard to a decision. It checks incoming data, shows the assumptions behind workload scenarios, and gives follow-up work an owner. The examples use synthetic data.

  • Validated measures
  • Explainable scenarios
  • Accountable follow-up
Open case note
Need
Bring separate trackers, dashboards, alerts, and case queues into a shared view where teams can find the definitions, supporting records, and person responsible.
What I built
A service-operations workspace with validation, process analysis, six-week backlog scenarios, owned cases, playbooks, and outcome tracking. Blocking defects remain visible but are excluded from trusted KPIs.
In practice
See what led to a finding, compare staffing assumptions, and assign the next step to someone who can follow it through.
Look inside
The synthetic walkthrough shows which records are excluded from KPIs and why. Six-week scenarios expose their assumptions, planning ranges, and held-out backtest, so a reader can examine how a recommendation was reached.

Public source v0.3.1 · Synthetic demonstration · Findings are associative, not causal proof · Demo authentication · External operational write-back disabled by default.

Explore the operations intelligence source
GOVERNED CATALOG / 05Public case study
DefinitionsOwnershipChecksImpact
05

Data governance · Lineage

Data Governance & Lineage Portal

Public case study · Private implementation

A governance case study for people who manage and use shared data. It connects business definitions to source fields, quality checks, and reports, helping a team see who owns the data and what a proposed change could affect.

  • Business-to-field context
  • Visible change impact
  • Explainable trust
Open case note
Need
Help owners, stewards, analysts, and technical teams answer the same questions about meaning, trust, ownership, and impact.
What I built
A local-first product that brings catalog records, glossary terms, classifications, quality controls, source fields, lineage, impact analysis, and permission-aware change review into one shared model.
In practice
Users can trace a governed concept from its definition and accountable people through source fields, transformations, consuming reports, quality checks, and recent changes.
Look inside
The public case study traces a business term through ownership, source fields, transformations, quality checks, and consuming reports. It shows how that context supports a change review; the implementation remains private.

Public case study · Version 0.3.0 historical qualification · Synthetic examples · Proprietary implementation and operational records remain private · Enterprise deployment is not claimed.

Review the governance portal case study

03 · How I work

Start with the people doing the work.

I start by understanding the job, the people doing it, and where things get stuck. Then I build around clear requirements, useful feedback, and a way to recover when something goes wrong.

Explore my approach

Confidence at
every step.

Trustworthy system operating model A four-step path connecting govern, observe, verify, and sustain. 01Govern 02Observe 03Verify 04Sustain

01 / Clear rules before automation

Agree on the information required, where it comes from, and who reviews it before work begins.

Why it mattersEveryone starts with shared expectations and clear responsibilities.

StandardsProvenanceOwnership

Working principles: governed, observable, adaptable, and human-centered.

04 · Experience

Builder range.
Operator discipline.

My career began in film and post-production, where the right information has to reach the right people at the right time. I bring that same care to data governance, reporting, and automation—connecting the details to the people who rely on them.

Oct 2026 — Present

Publicis Health Media · Manager, Data Enablement

My current role, bringing together my background in information management, analytics, and practical business delivery.

Sep 2018 — Present

Gateway Information Group · Consultant

Turn business needs into process maps, reporting, and automation using Python, PowerShell, Excel, Power BI, and APIs. Carry tools from requirements through testing and deployment, with attention to metadata, data quality, reliability, and privacy.

Feb 2022 — Jan 2025

Visa · Marketing Manager, Production Operations

Built Power BI dashboards and automated reporting for agency progress, spend, deliverables, and KPIs. Strengthened metadata and production workflows, and worked with governance stakeholders to bring responsible-AI expectations into agency onboarding and documentation.

Earlier career

Reporting, archive operations & production

Earlier roles at Advertising Production Resources, The Digital Archive Group, Amazon, Pangea Media Productions, and Harpo Productions shaped my approach to reconciliation, metadata, training, and coordinated delivery.

05 · Foundation

Creative roots.
Information rigor.

Creative production and information management come together in how I organize information, design clear experiences, and support the people using them.

2017 — 2019

M.S. in Information Management

Dominican University

Professional development

Governance & data practices

Information governance, data governance, and data-driven network security

2006 · Additional study

Film / Video & Post Production

Post Production Management · Columbia College Chicago

View résumé (PDF)

A small gift for curious builders

Two projects. One hands-on lesson.

I’m sharing these learning projects for builders who enjoy seeing an idea take shape. They explore complementary sides of a hypothetical trade: evaluating a 1¢ entry from a sample market snapshot and checking whether a partial sale covers supplied costs and fees. Both public projects are offline educational planners, so this is a learning pair—not a live-trading system.

01 / BUY Kalshi 10 × 1¢ Planner

Checks an explicitly synthetic market snapshot for exactly ten contracts at 1¢ each—10¢ before modeled fees. Outdated, incomplete, or conflicting inputs prevent a plan. The public source is offline, with no credential support or order authority.

Review the buy project 02 / SELL Kalshi 15-minute Sell Preview

An offline exit planner that checks whether a partial sale covers the full acquisition cost and modeled fees, valuing unsold contracts at zero. It models a 60% target with prices up to 2¢. Synthetic inputs only; no account access or live orders.

Explore the sell preview

Fixed-price teaching example · before fees

Model a partial exit. Keep the remaining position.

The retained fixed-2¢ example assumes ten hypothetical 1¢ acquisitions, all available, with no prior or reserved exits. Selling six would return 12¢ gross and leave four open. The current default also evaluates cost-backed prices up to 2¢; the 40% and 50% columns are comparisons, not selectable modes or expected returns.

40%8¢ gross6 remain

50%10¢ gross5 remain

60%12¢ gross4 remain

Educational, MIT-licensed projects—not a performance promise. Both projects are offline planning tools with no network client, credential support, or order authority. Examples assume completed 2¢ fills before fees. Liquidity, fees, and settlement affect results, and retained contracts remain at risk. Review current eligibility and platform terms before use. Independent projects; not affiliated with or endorsed by Kalshi.

06 · Start a conversation

Have a project in mind?

If your team is sorting out messy data, improving reporting, or finding a better way to get work done, I’d be glad to hear what you’re working on.

jerryrnapier@gmail.com
LinkedIn GitHub IMDb