Available now · freelance projects, contract work, and full-time roles

Turn operational drag into software people can inspect.

I build reliable Python pipelines, browser workflows, and internal tools for the repetitive work that falls between your existing systems.

Inspect selected work Start with a $350–$750 fix

Deterministic core Human approval Reproducible proof

release://portfolio/main

verified session

run release-check

typecheck0 errors0.42s

workflow tests12 passed1.18s

approval gatesprotected0.08s

production artifactverified2.07s

RELEASE READYdeterministic core · human approval intact

03verified project releases

12/12portfolio release checks

0actions hidden from the user

Selected systems / verified releases

Not concepts. Working systems with receipts.

Each case study separates what the release proves from what comes next—and keeps private or consequential actions outside the demo.

01 / Python data pipeline

Public bid research, normalized and reviewable.

TradeOps turns fragmented construction-opportunity pages into evidence-aware reports with provenance, duplicate prevention, trade classification, and source-rights gates.

Verified proof: Five synthetic records became four unique opportunities; one duplicate was collapsed and one source was quarantined.

Read the TradeOps Data case study

pipeline://opportunity-normalizationverified

INPUT05 records

TX-4401city portal

TX-4402county feed

TX-4402duplicate

TX-4403rights hold

TX-4404district page

01DEDUPE

02RIGHTS

03NORMALIZE

OUTPUT04 unique

4reviewable records

03released

01quarantined

02 / Chrome extension

Browser assistance that keeps the user in control.

A side-panel extension that captures the intended job tab, extracts structured details, drafts through a user-selected AI provider, and fills recognized non-sensitive fields for review.

Verified proof: 61 tests passed; the verified browser flow preserved the target, filled safe fields, excluded sensitive fields, and submitted nothing.

Read the ATS Copilot case study

jobs.example.test/applysynthetic fixture

Application4 fields detected

03 / Local-first internal tool

Messy CRM exports, made explainable.

A guided workflow for mapping, validating, deduplicating, prioritizing, auditing, and exporting CRM CSV data. Every merge or keep-separate decision requires confirmation.

Verified proof: The deterministic demo processed 24 synthetic records into 22 active records with all six duplicate candidates explicitly resolved.

Read the CRM Rescue case study

local://duplicate-reviewLOCAL FIRST

CRM Rescue duplicate review with two synthetic contact records and a required decision

verified artifact

Operating protocol

Inspect → Automate → Verify

Hover, focus, or tap each stage. The goal is not maximum automation. It is the smallest reliable system that makes the work easier to operate.

Active procedure

$ capture --workflow current-state

Source boundaries mapped
Failure states named
Approval owner identified

01signal locked

Two ways to work together

One painful workflow—or the whole team.

The same operating background supports both paths: understand the work, expose failure states, build the smallest useful system, and verify the result.

CLIENT_PROTOCOL / 01

For businesses

Fix one painful process first.

Start with a bounded spreadsheet, CSV, browser, or copy-and-paste workflow. The Automation Fix Sprint is fixed-scope, proof-first, and designed to create a useful result without a large commitment.

See the Automation Fix Sprint

TEAM_PROTOCOL / 02

For employers

Add an operator-minded builder.

I am open to remote roles involving Python automation, implementation, internal tools, technical operations, browser workflows, or data quality.

Discuss a role

Focused services

Automate the busywork. Keep the judgment.

Best for operational work that lives in spreadsheets, browser tabs, CSV exports, or repeated copy-and-paste.

01AVAILABLE MODULE

Workflow Automation Assessment and Quick Win

Map a repetitive process, identify the highest-value failure point, and deliver a bounded automation or implementation-ready roadmap.

02AVAILABLE MODULE

CRM/Data Cleanup and Operational Reporting

Turn inconsistent exports into validated, deduplicated, reviewable data and reproducible operational reports.

03AVAILABLE MODULE

Public-Data Monitoring or Browser Workflow Automation

Build careful collection and browser-assisted workflows with provenance, minimum permissions, and human approval gates.

See service boundaries and working principles

Ready for input

Bring me the process everyone works around.

One screenshot, sanitized sample, or short explanation is enough to start.

Start a conversation kori@tradeopsdata.com