High-Performance Engineering.
Heavy systems. Done right.

20 years in Enterprise systems (PayPal, Sage, Siemens). Ruby/Rails at scale, legacy modernization, and production-grade AI integration.

30 FTEs
Saved (Telstra)
Billions
Transactions handled (PayPal)
20 yrs
Enterprise Engineering
Start a conversation See case studies →
Selected work

Case Studies

Projects where the standard approach didn't work.

PayPal & FinTech

Billions of Transactions

Challenge: Migration and implementing changes in undocumented legacy code (Ruby/Microservices) at massive scale.

Engineering: Implementation of "percentage feature switch" for webhooks, enabling staged rollout with controlled blast radius.

Result: Production-safe deployment model — staged rollouts with controlled blast radius, full Splunk observability, organized service architecture.

Foot Locker

Black Friday

Challenge: Black Friday peak load on a large e-commerce platform.

Engineering: Multi-layer caching strategy implemented on our scope (the content/blog section) well ahead of peak.

Result: Our section absorbed Black Friday traffic without incident. While other parts of the platform faltered under load, the caching held.

ArsMacros · Telstra

Custom product, via IBM

Challenge: Thousands of hours lost to manual SharePoint/ITSM workflows on the client side.

Engineering: Built ArsMacros — our own automation product integrating and automating the data flow. Delivered to Telstra through IBM channel partnership.

Result: Manual workload equivalent to ~30 FTEs replaced by automation. ROI measured in millions.

Lab notes

Currently exploring

Where I'm spending non-billable hours — patterns I bring back into client engagements.

// 01

Agentic coding in production

Patterns for running Claude Code and similar agents against enterprise codebases — moving beyond demos to systems that actually ship. What's reliable, what isn't, what to wrap.

// 02

Durable workflow design

Deterministic skeletons with AI-controlled steps inside. Workflows you can audit, rerun, and trust — where "agentic" meets reproducibility.

// 03

Context engineering for RAG

Building retrieval systems that don't hallucinate. Source provenance, deterministic output, guardrails for domains where being wrong is expensive.

Also currently

Learning wet lab biology — PCR techniques and bioinformatics, certified at the University of Gdańsk. The intersection of biology and AI agents is what makes GeneQuest possible.

Open to commissioned R&D in these areas. Get in touch →

Contact

I reply personally. Usually within 24h.

  • @ dm@humantechnology.com.pl
  • tel +48 884 160 776
  • loc Zielona Góra / Remote

Past experience:

PayPal Siemens Sage Foot Locker