Data Infrastructure · Automation · Operations — Chicago / DuPage County, IL

Stop doing things by hand.

I'm Devin Oommen. I help small and mid-sized businesses replace manual operations — paper counts, scattered spreadsheets, systems that don't talk — with software that runs itself. I clean the data, build the tool, and make sure it holds up in production.

2 days → 30 min
Manual prep eliminated
30+ hrs/wk
Recovered from manual data work
10 → 1
Disconnected systems unified
6.4M+
Rows in production pipelines
Offline-first
Field tools deployed to production
Services

What I do

Your data and your day-to-day should work for you, not the other way around.

AI & Document Intelligence

Turn documents, emails, and recorded meetings into structured data, summaries, and action items. AI that works off your own business data: pull the numbers out of PDFs and invoices, answer plain-English questions over your records, and turn every meeting into a clean list of decisions and to-dos.

Operations & Automation

Inventory counts, email workflows, scheduling, weekly reports — automated. Your team stops re-keying numbers between spreadsheets.

Dashboards & Reporting

One dashboard instead of five spreadsheets and a gut feeling. KPI reports that update themselves, so you always know where the business stands.

Custom Internal Tools

Software built for how your business actually works — phone apps, desktop tools, manager dashboards. No SaaS subscription, no forcing your process into someone else's box.

Data Cleanup & Integration

Messy records, duplicate data, systems that don't talk? I clean it, connect it, and turn it into one source of truth you can actually decide from.

Web Scraping & Data Collection

The data you need is online but has no download button — competitor pricing, public records, government sites. I build collectors that gather and structure it for you.

Civic & Public-Records Data

Large-scale public-records pipelines, turnout and targeting models, and civic data platforms for campaigns, newsrooms, and researchers.

Selected work

Real engagements, and what changed

What these systems replaced, and what they delivered. Client identities kept private.

Operations & Automation

Paper inventory → a system that runs itself

A multi-location specialty coffee roaster · ongoing
The problem

Staff counted inventory on paper and re-keyed it by hand every week.

What I built

An offline-first inventory app employees install on their phones. Every count writes to both a permanent ledger and the live ordering grid, with automatic weekly rollover, one-click setup for new locations, and a manager command-center dashboard. Plus email automation across the company's accounts and a searchable employee manual built from 16 source documents.

The result

Paper counts replaced with a phone tap · weekly rollover automated · live across locations.

Operations Platform

10 disconnected systems → one platform

A multi-location medical practice
The problem

The practice ran on 10 separate systems. Prepping a single patient-day took two full days of manual work.

What I built

One platform for scheduling, check-in, KPI tracking, and daily reports. ETL pipelines pull live data straight from their practice-management software into a 30+ metric reporting framework.

The result

2 days of manual prep → 30 minutes · ~30+ hours/week of manual data collection eliminated.

Data Platform

A 6.4-million-row public data platform

Follow The Money IL · my own project
Visit the live site ↗
The problem

Illinois and federal campaign-finance data is public but nearly impossible to actually search.

What I built

A platform that ingests state filings and federal FEC data on scheduled syncs and makes all of it searchable, with donor and network analytics on top.

The result

6.4M+ receipt rows · 530K+ filing totals · 1.3M+ donor-analytics rows · live in production.

Also built
Legislative Bill Stats — full-stack civic dashboard (live ↗) Board-of-trustees financial ETL pipeline for academic research Honors thesis: statistical analysis of 484,630 villages Monte Carlo robustness study (ANOVA)
About

The person behind the systems

I'm Devin Oommen. I build software and data infrastructure for businesses that have outgrown spreadsheets but don't want a full engineering department.

My background: graduate training in statistics and econometrics, years shipping production software, and a stretch as the final QA gate on an FDA-cleared medical device used in 100+ hospitals. I can clean your data, model it, build the tool that uses it, and make sure it holds up when people depend on it.

I keep my client list short. My anchor engagement is an ongoing partnership with a multi-location specialty coffee roaster, and I take on a few clients at a time, mostly around Chicago and DuPage County. If you're sitting on manual processes or data you can't use, let's talk.

Tools & technologies
PythonRSQLJavaScript pandas / NumPyscikit-learn / LightGBMFastAPI / Flask Power Apps / Power Query / DataverseGoogle Apps ScriptSQLite ETL pipelinesPlaywright / Selenium / BeautifulSoupPDF parsing Whisper STTGeospatial / LeafletTableauAWS / AzureGit
Graduate-level econometrics & computational statistics · MPSA conference presenter · FDA/EU device QA · multiple journalism & public-service awards.

Let's talk about your operations.

Tell me your biggest pain point and I'll tell you honestly whether I can help. Free 30-minute call, no pitch.

No spam, ever — your details come straight to me.

Prefer email? devin@oommen.org · call or text 630-456-6784