#ai#machinelearning#testing#python
I looked inside a small AI model to find where luck changes its answer
In my last article, AI judges scored the same decision lower when it happened to end in harm. I...
Sara Bezjak5 min read
Author profile
Sara Bezjak
Software & QA engineer testing AI - where to trust it and where to catch it.
Articles by @sara_bezjak
Browse the latest writing surfaced through DevArt.
Sara Bezjak5 min read
#ai#testing#llm#claude
Your AI judge might be grading luck, not decisions
A support agent is about to email 8,412 people. Its instructions say to send a test email first, and...
Sara Bezjak11 min read
#ai#python#finetuning#machinelearning
Fine-tuning a content-moderation model on a laptop is easy. Trusting the result is not.
I fine-tuned a small language model to sort messages into safe or unsafe, the kind of check a...
Sara Bezjak8 min read
#ai#qa#testing#python
The model said it read the report. It didn't.
I'm a regular automation engineer. Over a few months I built five small projects that test local AI...
Sara Bezjak8 min read
#ai#testing#llm#python
A benchmark is only as good as the model you use to grade it
I built a pytest harness that runs the same set of questions through five language models at once - a...
Sara Bezjak9 min read
#agents#ai#security#testing
With an AI agent, the answer is the last place the bug shows up
I built a pytest suite for a small AI agent - not a model that answers once, but one that plans,...
Sara Bezjak7 min read
#ai#security#testing#python
The hard part of attacking an AI isn't breaking it. It's telling real harm from fake.
I built a red-team test suite that fires adversarial prompts at an LLM-backed API and decides, for...
Sara Bezjak7 min read
#ai#python#testing#rag
Search bug or model bug - testing a RAG system to tell them apart
I'm an automation tester. Usually my job is simple: the same input should give the same output, every...
Sara Bezjak8 min read
#flutter#androiddev#beginners#mobile
From a spreadsheet of questions to an app on Google Play
Origins I've always been drawn to deeper questions. I'd collected a few notebooks with...
Sara Bezjak4 min read
#ai#testing#python#qa
Five ways to test an LLM's answer and what each one misses
I'm a regular automation engineer. My usual job is checking that an app does the same thing every...
Sara Bezjak6 min read
#ai#testing#python#qa
A QA engineer's first AI testing project - FastAPI + local LLM + pytest
What testing an LLM taught me that UI testing didn't. A small FastAPI + Ollama + pytest project, the bugs I hit, and what clicked.
Sara Bezjak5 min read
AI Tools for Existing Playwright + Pytest Frameworks: What Actually Works
Purpose Research and evaluate AI-powered tools and workflows to improve test automation...
Sara Bezjak4 min readAdvertisement