I Am Bad At Things

Presented at PyCon AU 2026 in Brisbane, QLD

Transcript

1. I am bad at things And that's okay PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 1
2. Noah Kantrowitz • • • • • He/him coderanger.net | cloudisland.nz/@coderanger Kubernetes and Python SRE/Platform for Geomagical Labs, part of IKEA We do CV/AR for the home PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 2
3. I Am A Squishy Human I have made mistakes before I will make mistakes again This is okay This must be okay PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 3
4. Building Resilient Systems Technical Resilience Attack Resilience Human Resilience PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 4
5. Processes Assume Failure Be more than the sum of our parts This is not (just) about "AI" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 5
6. 1. How Do We Fail? 2. Where Can We Improve? 3. What Is The Human Cost? PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 6
7. Ways Humans Are Fallible We are predictable Named biases are lenses PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 7
8. Confirmation Bias Selective perception Seeing what we want to see The magic seven words "You are already doing the right thing" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 8
9. Confirmation Bias I am good at this People who are good at this do the right thing I must be doing the right thing PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 9
10. Confirmation Bias Authority Bias "Well the senior engineer said to" "It's standard practice" Social Cohesion and Conformity "I don't want to be one of those reviewers" "We can always fix it later" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 10
11. Inattentional Blindness "Did you notice the gorilla?" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 11
12. Recency Bias "Everyone drop what you're doing to fix today's bug" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 12
13. Repetition Blindness "I have to go to the the store every evening" x += 1 x += 2 x =+ 3 PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 13
14. Overconfidence Bias Better-than-average effect Planning fallacy Overprecision Aside: Dunning–Kruger is bad science PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 14
15. Where Can We Improve? PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 15
16. Checklist Culture Plan your dive and dive your plan Thinking when not in the moment Following when stressed PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 16
17. Types of Checklists Merge request Deploy Incident response ... PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 17
18. 4 Signs You May Need A Checklist 1. Frequent tasks 2. Repetitive tasks 3. High stress, usually time sensitive 4. High impact of errors PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 18
19. Checklists Must Be Followable PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 19
20. Checklists Must Be Followable Checklists Must Be Practiced PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 20
21. Checklists Must Be Followable Checklists Must Be Practiced Checklists Must Be Blameless PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 21
22. Objective vs Subjective Objectivity reduces bias Everything is a spectrum PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 22
23. Unit Tests! PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 23
24. Unit Tests! Are Subjective assert True is True PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 24
25. Coverage Reporting What are we actually testing? "100% coverage required" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 25
26. Patch Coverage It's great, use it Try Codecov if you can self-host PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 26
27. "You didn't write enough tests" "This line is important and isn't covered" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 27
28. Static Typing return a.get("b") + 1 TypeError: unsupported operand types PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 28
29. A QA Engineer Walks Into A Bar They order a drink They order -1 They order None They order "one" They order [1, 2, 3] PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 29
30. Stylistic Consistency You can have any color as long as it's Black PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 30
31. Consistency is Maintainability Flake8 Black isort Ruff PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 31
32. Add Your Own [project.entry-points."flake8.extension"] X = "flake8_example:ExamplePlugin" Have opinions! CharField(null=True, blank=True) # vs. CharField(blank=True, null=True) PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 32
33. Code Review We do still need humans Authors count too AI will not save you PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 33
34. Kinds of Review "Flight attendants please arm doors and cross-check" "Does this SQL query look correct before I run it?" "What does rm -rf * mean?" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 34
35. What Is Code Review For? Reducing defects Team training Architecture consensus Specialist consultation PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 35
36. What's Wrong With Code Review? Easy to miss things Time consuming Tiring PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 36
37. Micro vs Macro Review • • Micro - Looking for self-contained problems • "You used the wrong variable name" • "This library isn't thread safe" • "We have a utility function for this already" Macro - Looking at overall structure • "How does this handle authentication?" • "What will the performance of this look like under load?" • "This doesn't match our usual pattern for this kind of API" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 37
38. Nits or Bikesheds Bad Logic or Design Missing Elements PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 38
39. Macro Review Doesn't Work 1. Read the diff, mostly focusing on "vibes" as you go 2. Use the diff to reconstruct the mental state of the author 3. Imagine what your mental state would be when doing the same work 4. Compare that extrapolated mental state to your assumptions 5. If there are mismatches, highlight them for further reading PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 39
40. Reviewing Mental State What were you thinking when you wrote this? PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 40
41. Aside: AI Code Review Humans have "non-local consistency" Machines do not have a "mental" Or any states thereof PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 41
42. You Should Do It Anyway Again: lots of side benefits Cross-training, education, discussion Rubber ducks can still find bugs Just don't assume you'll find all the problems PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 42
43. "These new tools are fine because we'll just notice if they do the wrong thing" PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 43
44. Hypervigilance and Burnout PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 44
45. I Am Not A Mental Health Expert Hitting the Wall and How to Get Up Again Tackling Burnout and Strategies for Self Care Jackson Fairchild – PyCon AU 2016 PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 45
46. Cognitive Load Humans have a RAM limit Or spell slots if you are a D&D fan PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 46
47. Examples of Loops Did we validate every TLS cert Are all function names consistent Have I forgotten any code reviews PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 47
48. Cognitive Offload Shift to machines Shift to AI (if it actually helps) Shift to colleagues PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 48
49. Fixing Things • • • • • • Understand common cognitive biases Supplement human decision making with checklists Look for objective, deterministic guardrails Build review processes that assume failures Cognitive offload, but use your powers only for good Fix processes, not people PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 49
50. Be Kind To Yourself PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 50
51. Questions? PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 51
52. • • Building resilient systems • Gesture in the direction of technical resilience but not this talk • Human resilience, building systems that can survive human fallibility • Also not a talk about blue team, malicious humans are a different kind of threat • We all make mistakes, how can we build systems which are better than any one us individually • Sections: • How we fail • Where can we improve • What is the human cost Ways humans are fallible • Confirmation bias • Selective perception PyCon AU 2026 – Noah Kantrowitz – @coderanger@cloudisland.nz 52