Ines Panker

Hi, I'm Ines Panker.

I'm a senior staff engineer at Sleuth, a SaaS observability platform (the first engineer they hired, de-facto architect), and a conference speaker on the human side of engineering, from Slovenia.

I take complicated systems and find the arrangement that makes them work like poetry.

Almost two decades of building software.

Multi-tenant architecture. Circuit breakers guarding over a million API calls a day. Real-time anomaly detection for hundreds of clients. An LLM pipeline, a RAG system, and a running experiment: how much of my own work can LLM agents carry.

Python Go Django Celery PostgreSQL Redis RabbitMQ Elasticsearch LLM agents Conference speaking

How to contact me: LinkedIn works best, or catch me at one of the conferences below.

More about me My CV My talks

Latest post 23. September 2026 · API Resilience Patterns Pattern #4: Load Shedding, or Letting Tasks Expire on PurposeLoad shedding is the deliberate dropping of some work when your system can’t keep up. It’s about choosing which work to do and which to reject, even though the general...

What I do

Build software

Almost two decades of writing code: Java first, then PHP, then a decade of Python. Backends, data models, the architecture around them.

Untangle systems

I take codebases that have outgrown their teams and make them understandable again. Like reverse-engineering the undocumented data layer of a compliance platform and rebuilding its model layer.

Speak & teach

Lots of conference talks across Europe, all on my own time: software estimation, detecting bad deploys with math, API resilience, cognitive biases in code.

Speaking

I speak in my free time: no sponsors, no employer agenda, just topics I care about.

  • Dec 2026 BuildStuff 2026 Did My Deploy Degrade Production? Detecting Bad Deploys with Math, Not AI
  • Nov 2026 Heapcon 2026 Software Estimates and the Illusion of Control
  • Oct 2026 TDC 2026 Did My Deploy Degrade Production? Detecting Bad Deploys with Math
  • Oct 2026 Devoxx Belgium 2026 Their API Went Down. So Why Is My App on Fire?
  • Sep 2026 Python Meetup Ljubljana Their API Went Down, So Why is My App on Fire?

All talks and the talk catalogue

Selected writing

  • API Resilience Patterns

    Latest Pattern #4: Load Shedding, or Letting Tasks Expire on Purpose

    Load shedding is the deliberate dropping of some work when your system can’t keep up. It’s about choosing which work to do and which to reject, even though the general...

    API Resilience PatternsCode Patterns

  • API Resilience Patterns · a series of 5 posts

    External APIs are a pain to integrate… particularly because they each follow their own rules, but also because these rules are often badly documented. It could be so simple: authenticate,...

  • The Algorithm Decided · a series of 1 posts

    Of course, we learned statistics. We went to university, there was a course, there was an exam, we passed it. But we don’t really use it at work… when writing...

  • Celery, the Python Task Queue · a topic of 4 posts

    Celery has been my daily companion for >5years now. Somehow our setup was never simple enough that we would have been allowed to just set it and forget it. But,...

  • Detecting Bad Deploys with Math, Not AI

    When you deploy some code, do you know its impact on your production stats, like memory usage, request duration, error rates? If you wanted to know if your code improved...

    Algorithms

  • Self-Documenting Code · a series of 4 posts

    What do we want? All the documentation! Are we going to read it? Only sometimes! So, you need to be sneaky! The only reliable way to get other devs to...

  • Why do smart devs write silly code? · a series of 3 posts

    Psychology has studied cognitive biases for decades, but in software engineering we rarely talk about them, even though they affect everything from architecture to debugging to reviewing. We make decisions...

  • Software estimation is a delusion

    You will never be able to accurately estimate the time and effort it will take you to build a piece of software .. for as long as you keep doing...

    Engineering Research

Archive of all posts