Caleb Fedyshen

Welcome to my blogthing – I heard all the best programmers have the most retro looking websites. There are many secrets on this site, many more on the desktop version of the site than mobile. Have fun! I hope you're having a great day.
About Caleb

I replace overpriced enterprise software with fast custom apps that pay for themselves.

Designer/builder of tools, data pipelines, and automations that remove manual work and slash license costs. In one engagement, I replaced a legacy banking image viewer and saved over $1.14M CAD per year. If you’re spending big on clunky software—or spending people’s time to work around it, I can fix that.

Proof

$1.14M+/yr
license spend eliminated
1000s
of hours saved across several businesses
5+
apps delivered
3.9 GPA
4× Dean’s List, UCalgary CS
Hundreds
of employees used my Teams-integrated app
Dozens
of processes automated end-to-end
Signature Case Study

Killing a $70,000/month tool with a lean replacement

A retail bank’s “Image Viewer” let teams retrieve check images. It was being sunset and the replacement was usage-priced—forecasted at $70k USD/month for Plains. Treasury needed speed, accuracy, and zero vendor lock-in.

What I built

  • End-to-end system: multithreaded extraction of terabytes from thousands of zips;
  • Parsed millions of XML entries into a queryable database with incremental refresh;
  • A fast Flask web app deployed as Databricks App—search, preview, export, audit trails;
  • Automated the manual steps the legacy tool couldn’t.

Outcome

  • $70,000 USD/month saved vs. the bank’s new pricing (~$1.15M CAD/year).
  • Used by finance leadership; faster retrieval, fewer steps, better visibility.
PythonFlaskDatabricks AppsAzureSQLHTML/CSS/JS

What I do

  • Replace expensive licenses with focused internal tools.
  • Build data products—ingestion, transformation, search, and friendly UIs.
  • Automate workflows end to end across Azure, Databricks, Power Platform.
  • Make dashboards useful (not just pretty) with the right operational hooks.
PythonFlaskDatabricksAzure Data FactoryPowerAppsSQL

How engagement works

  1. Fit call (15–30 min): we verify there’s a path to ROI.
  2. One-week sprint: clickable prototype or a live slice in prod.
  3. Scale-up: productionize, docs, handoff, and support options.

You keep the code. Security & access handled your way.

Selected projects

  • Blaze Invoicing — full-stack invoicing for an Alberta accounting firm (Supabase, Flask, DigitalOcean).
  • Snow Ops — task assignment + status for a snow removal company (internal tool).
  • Green Energy Visualizer — interactive biodigester flow in Godot; ground-up art + animation.

Toolbox

Languages: Python, Java, C++, ARMv8, SQL, Haskell, HTML/CSS

Platforms: Azure (Databricks, ADF, Data Lake, Fabric SQL), Power Platform, DigitalOcean

Soft skills: Safety mindset, teaching, clear comms, relentless problem-solving

Plains MidstreamSuncorHyprPixlCalgaryToSpace

A bit about me

Calgary based software developer and Data Architect finishing a B.Sc. in Computer Science at the University of Calgary (3.9 GPA, 4× Dean’s List). I like turning messy, slow processes into crisp systems that make teams faster.

I build for the real world: reliability first, aesthetics included.

FAQ

Why custom vs. off-the-shelf?

Because you pay for a thousand features you’ll never use, and still do manual work. I build the exact thing your process needs, integrate it with your stack, and prove ROI in weeks, not years.

Where do you deploy?

Your cloud, your rules. Databricks Apps, Azure, or a simple VM—whatever fits your governance and security posture.

How do we start?

Email me a short description of the problem and any screenshots. I’ll outline a path to ROI and a first sprint plan.