Dimitris Kyrkos

11

4 min read

A clean paper trail never guarantees safe code

Dimitris Kyrkos

Nov 5

Your build attestation is a chain of custody, not a safety certificate

#discuss #security #ai #devops

13

16

4 min read

Shared blind spots hide simple logic flaws

Dimitris Kyrkos

Nov 2

Your AI code reviewer isn't a second opinion. It may be the first opinion again.

#discuss #codereview #security #ai

27

40

4 min read

Replacing audit trails with non-deterministic text

Dimitris Kyrkos

Oct 29

An AI summary is not a fix. Stop letting it close your security findings.

#discuss #security #ai #codereview

39

75

3 min read

Debug pastes leak creds past git

Dimitris Kyrkos

Oct 26

Your secrets don't need a git commit to leak into an AI tool

#discuss #security #ai #programming

75

137

3 min read

Uncalibrated model scores create automated risk

Dimitris Kyrkos

Oct 22

"I'm 93% confident" is a string, not a probability

#discuss #ai #machinelearning #programming

34

39

3 min read

Comments highlight failing third-party APIs

Dimitris Kyrkos

Oct 19

Your agent's retry bug isn't an AI problem. It's a distributed-systems problem with an LLM on top.

#discuss #ai #architecture #programming

74

169

3 min read

Dimitris Kyrkos

Oct 15

Your model router is a classifier, and you probably aren't evaluating it like one

#discuss #ai #programming #machinelearning

10

3 min read

Dimitris Kyrkos

Oct 12

Human-in-the-loop isn't a safety net. It's a UX problem with a compliance label.

#discuss #ai #machinelearning #ux

3 min read

Metrics miss maintenance debt

Dimitris Kyrkos

Oct 8

Shipping faster with AI isn't engineering maturity. It's a demo that hasn't met year two yet.

#discuss #ai #architecture #programming

20

23

3 min read

Dimitris Kyrkos

Oct 5

Your bugs used to burn CPU. Now they burn money.

#discuss #ai #architecture #webdev

25

3 min read

Dimitris Kyrkos

Oct 1

Your AI feature isn't a feature. It's a dependency you don't control.

#discuss #ai #architecture #webdev

18

3 min read

Real-world examples like regex-replacing prompts

Dimitris Kyrkos

Sep 28

Half the AI agents in production are if-statements with a GPU bill

#discuss #ai #programming #architecture

46

99

3 min read

Dimitris Kyrkos

Sep 24

Your model doesn't need more training. It needs a better search index.

#discuss #ai #machinelearning #architecture

24

4 min read

How summarization breaks multi-step workflows

Dimitris Kyrkos

Sep 21

Your LLM has no memory. Your application had better have one.

#discuss #ai #architecture #programming

24

36

4 min read

Dimitris Kyrkos

Sep 18

Two-second latency isn't an AI problem. It's an architecture problem your stack was never built to hide.

#discuss #ai #webdev #programming

11

3 min read

The hidden danger of 90% accuracy in production

Dimitris Kyrkos

Sep 15

Your LLM Isn't Bad At Math. It Was Never Doing Math In The First Place.

#discuss #ai #architecture #programming

23

4 min read

Dimitris Kyrkos

Sep 11

AI Model Upgrades Aren't Patches. They're Silent Breaking Changes Wearing a Version Bump.

#discuss #ai #architecture #programming

4 min read

Dimitris Kyrkos

Sep 8

AI Didn't Kill the Need for System Design. It Just Made Bad System Design Easier to Ship.

#discuss #ai #architecture #programming

30

12

4 min read

Commenters share how wrong algorithms still pass

Dimitris Kyrkos

Sep 4

Your AI-generated tests aren't testing your code. They're testing the AI's blind spots.

#discuss #ai #programming #software

40

53

3 min read

Dimitris Kyrkos

Sep 1

Semantic caching isn't a cost-saving hack. It's an admission that most "AI features" are FAQ bots in disguise.

#discuss #ai #architecture #programming

3 min read

Dimitris Kyrkos

Aug 28

AI autocomplete isn't a productivity tool. It's a judgment test you take every few seconds.

#discuss #ai #programming #webdev

16

3 min read

Dimitris Kyrkos

Aug 25

AI-Generated Code Doesn't Have a Readability Problem. It Has a Style-Enforcement Problem Wearing an AI Hat.

#discuss #ai #programming #software

3 min read

Pragmatic gut checks for merging AI code

Dimitris Kyrkos

Aug 17

Using AI to Code Isn't the Risk. Not Understanding What It Shipped Is

#discuss #ai #programming #software

12

3 min read

Dimitris Kyrkos

Aug 4

Your model doesn't need to pass the bar exam. It needs to parse a log file.

#discuss #ai #architecture #programming

24

25

3 min read

Dimitris Kyrkos

Jul 31

The all-purpose agent isn't an architecture. It's a single point of failure with a system prompt.

#discuss #ai #devops #programming

21

3 min read

Production latency and token cost pitfalls

Dimitris Kyrkos

Jul 27

"Unlimited context" is not a feature. It's technical debt with better marketing.

#discuss #ai #architecture #programming

40

36

3 min read

Dimitris Kyrkos

Jul 23

Banning ChatGPT Won't Fix Your Shadow AI Problem. It'll Just Hide It.

#discuss #ai #security #devops

20

3 min read

Dimitris Kyrkos

Jul 21

RAG isn't an AI problem. It's a data engineering problem wearing an AI hat.

#discuss #ai #programming #webdev

30

4 min read

Dimitris Kyrkos

Jul 17

The fallacy of "AI-first." Start with the friction, not the technology.

#discuss #ai #programming #webdev

30

3 min read

Dimitris Kyrkos

Jul 14

AI frameworks make the first 10% feel like magic. The other 90% is where they break you.

#discuss #ai #programming #webdev

19

3 min read

Dimitris Kyrkos

Jul 7

The AI conversation is shifting from "what can it do" to "can we rely on it"

#discuss #ai #webdev #programming

34

25

3 min read

Dimitris Kyrkos

Jul 3

The biggest barrier to enterprise AI adoption isn't the model. It's trust in everything around it.

#discuss #ai #programming #webdev

3 min read

Dimitris Kyrkos

Jul 1

Attackers are hijacking exposed AI endpoints to run offensive operations. No exploit needed.

#discuss #security #ai #devops

16

4 min read

Dimitris Kyrkos

Jun 29

5 Even More Advanced Java Tips That Senior Engineers Actually Use

#java #programming #performance #tutorial

15

9 min read

AI skipping critical requirement friction

Dimitris Kyrkos

Jun 26

Functional doesn't mean correct. That's the biggest risk with AI-generated code.

#discuss #programming #ai #webdev

29

67

3 min read

Dimitris Kyrkos

Jun 22

AI isn't a software upgrade. It's an organizational redesign.

#discuss #ai #programming #webdev

12

2 min read

The widening gap between code and context

Dimitris Kyrkos

Jun 19

AI makes writing code easier. It doesn't make engineering easier.

#discuss #programming #ai #webdev

29

34

2 min read

Dimitris Kyrkos

Jun 18

5 More Advanced Java Tips That Senior Engineers Actually Use

#java #programming #performance #tutorial

8

9 min read

Dimitris Kyrkos

Jun 16

The Novo Nordisk breach hit patient data and proprietary AI models. The attack surface is expanding.

#discuss #security #ai #devops

10

3 min read

Dimitris Kyrkos

Jun 12

5 Advanced Java Tips That Senior Engineers Actually Use

#java #programming #performance #tutorial

6 min read

Dimitris Kyrkos

Jun 11

Vibe coding should accelerate development, not bypass validation

#discuss #ai #programming #webdev

8

2 min read

Dimitris Kyrkos

Jun 9

AI's real value isn't automation. It's how fast you can act on what you already know.

#discuss #ai #programming #devops

2 min read

Dimitris Kyrkos

Jun 8

2026's breach list is out and the pattern is depressingly familiar

#discuss #security #programming #devops

2 min read

Dimitris Kyrkos

Jun 4

Vibe coding gets you the prototype. Then what?

#discuss #programming #ai #webdev

2 min read

Dimitris Kyrkos

Jun 3

AI projects don't fail because the model is bad. They fail because nobody planned for the mess around it.

#discuss #ai #devops #programming

1 min read

Dimitris Kyrkos

May 29

LLMs Generate Code, But They Can't Absorb Accountability

#discuss #ai #programming #webdev

4 min read

Dimitris Kyrkos

May 28

185,000 Affected in 7-Eleven Breach: Why Salesforce Is the New Soft Target for ShinyHunters

#discuss #security #webdev #programming

4 min read

Dimitris Kyrkos

May 26

Most Organizations Don't Have an AI Problem, They Have an Integration Problem

#discuss #ai #programming #webdev

8

4 min read

Dimitris Kyrkos

May 25

The GitHub Breach Is a Wake-Up Call: Your Dev Toolchain Is an Attack Surface

#discuss #security #github #devops

3 min read

Dimitris Kyrkos

May 19

Vibe Coding Is Great for Starting, But Not for Knowing

#discuss #webdev #programming #ai

5 min read

Dimitris Kyrkos

May 15

The OpenAI Breach Wasn't About OpenAI – It Was About the 84 Packages Above Them

#security #opensource #devops #webdev

2 min read

Dimitris Kyrkos

May 12

Why Your AI Product Breaks in Production: It's a Distributed Systems Problem, Not a Model Problem

#discuss #webdev #programming #ai

4 min read

Dimitris Kyrkos

May 11

Why Most AI-Generated Codebases Accumulate Invisible Technical Debt Faster Than Expected

#discuss #ai #webdev #programming

5 min read

Dimitris Kyrkos

May 8

State-Sponsored Hackers Are Exploiting Palo Alto Firewalls Right Now – And There's No Patch Yet

#security #cybersecurity #webdev #programming

3 min read

Dimitris Kyrkos

May 7

2.45 Billion Requests, 1.2 Million IPs: Why Traditional Rate Limiting Is Dead

#discuss #cybersecurity #programming #webdev

4 min read

Dimitris Kyrkos

May 6

When the Platform Your School Trusts Gets Hacked, Who's Actually Responsible?

#discuss #security #ai #cybersecurity

4 min read

Dimitris Kyrkos

May 5

Why Most AI Developer Tools Fail (It's Not What You Think)

#ai #devops #productivity #webdev

4 min read

Dimitris Kyrkos

Apr 29

AI Guardrails in Production: The Boring Engineering That Makes AI Features Actually Work

#ai #devops #code #programming

6 min read

Dimitris Kyrkos

Apr 24

"Beyond Linting: A Data-Driven Approach to Suggesting Better Code, Not Just Flagging Bad Code"

#codequality #machinelearning #softwareengineering #tooling