Latest posts tagged with #AIQuality on Bluesky
Sales Qualification Agent: How we evaluated and improved AI quality with benchmarks: Sales Qualification Agent (SQA) is not a simple productivity tool—it is a complex multi-step agent directly influencing revenue outcomes. The Sales… @MSFTDynamics365 #SalesQualification #AIQuality #Dynamic365
Remember SQL Injection? Simple times.
Now we have Prompt Injection. The art of convincing your AI to ignore instructions.
From buying a car for $1 to pirate jokes - it sounds funny until it happens to you.
Start thinking like attackers.
youtu.be/vc-rJifDBM4
#PromptInjection #AIQuality
The "Accept All" button is the new "I have read the terms and conditions."
AI spits out 200 lines of code, and we move on. But if no one is reviewing the slop, who actually owns the system?
The Maintenance Wall is closer than you think.
youtu.be/AU3XPz2l3L8
#AIQuality #SDET #TechnicalDebt
The "Accept All" button is the new "I have read the terms and conditions."
AI spits out 200 lines of code, and we move on. But if no one is reviewing the slop, who actually owns the system?
The Maintenance Wall is closer than you think.
youtu.be/AU3XPz2l3L8
#AIQuality #SoftwareEngineering #SDET
In the AI world, a bug isn't just "wrong behavior". It can be dangerous. We’re talking about real-world consequences and safety risks.
How do you test for the stuff that should never happen?
New video is live:
youtu.be/FrJ9FtSNMJs
#AIQuality #SoftwareTesting #TestinGil
#AIQuality #AITesting #ModelValidation #DataPipelineTesting #BiasDetection #DriftMonitoring #ScalableAI #StressTesting #QAAutomation #ResponsibleAI
We survived the monsters. Yesterday's "Anatomy of AI Quality" webinar was a blast.
Key lesson: Quality is the engine for AI speed. You can't ship sustainably without governance.
Opening 3 slots for the AI-Risk Diagnostic. Closes Monday!
testingil.com/applying-for...
#AIQuality #SoftwareTesting
Chen says AI quality isn’t a set‑and‑forget game—it's all about constant experiments, tweaking metrics, and polishing answer completeness. Curious how generative models stay sharp? Dive in. #AIQuality #EvaluationMetrics #Experimentation
🔗 aidailypost.com/news/chen-sa...
Testing AI with traditional techniques is like waving a flashlight in a dark forest.
You see a patch of green, but you're blind to the cliff 5 feet away.
We need a better way to watch for monsters.
Watch:
youtu.be/mLy-dHcu9Ik
#AIQuality #SoftwareTesting #SoftwareStrategy #ResponsibleAI
Testing AI with traditional techniques is like waving a flashlight in a dark forest.
You see a patch of green, but you're blind to the cliff 5 feet away.
We need a better way to watch for monsters.
Watch:
youtu.be/mLy-dHcu9Ik
#AIQuality #SoftwareTesting #EngineeringLeadershipl #SoftwareStrategy
Regular quality methods we used are not enough any more.
What we need is a holistic quality picture. We need to see a lot more to make better decisions.
Join the Webinar: The Anatomy of AI Quality
Here's a preview.
youtu.be/mZoeYuPlSrg
#AIQuality #SoftwareEngineering #TestinGil #QualityAssurance
Ensuring the quality of AI-generated code is paramount. Can AI consistently produce robust, secure, and maintainable solutions? Rigorous review and testing by humans remain critical for production readiness. #AIQuality 5/7
In the AI world, a bug isn't just "wrong behavior".
It can be dangerous. We’re talking about real-world consequences and safety risks.
How do you test for the stuff that should never happen?
New video is live:
youtu.be/FrJ9FtSNMJs
#AIQuality #SoftwareTesting#TestinGil #QualityEngineering
AI performance bugs hit us in two places: The user experience and our wallet.
UX we understand. We don't like to wait.
But the Invisible Tax? It can cost us millions.
Watch the full breakdown here:
youtu.be/XrE8Dwgb3xw
#AIQuality #PerformanceTesting #SoftwareEngineering #FinOps
AI performance bugs hit us in two places: The user experience and our wallet.
UX we understand. We don't like to wait.
But the Invisible Tax? It can cost us millions.
Watch the full breakdown here:
youtu.be/XrE8Dwgb3xw
#AIQuality #PerformanceTesting #SoftwareEngineering #FinOps
"This AI bug behavior is weird."
Tell that to your CTO and they hear ""this guy doesn't understand what they're doing".
We should report bugs like experts, not like ghost chasers. That means speaking differently.
testingil.com/2026/01/spea...
#AIQuality #SoftwareTesting #Bugs #BugReporting
Met some friends building AI apps. The consensus? Teams skipping the right way are about to hit a wall.
AI is complex, but the fundamentals haven't changed: Testing, CI/CD, and real engineering.
DM for a Q1 Quality Audit.
#AIQuality #TestinGil #QualityAssurance
My car just complained. Bleep. The mechanic then has to go in there.
What if it gave a diagnostic printout instead? The fix is faster and cheaper.
I'll show you how to build that "printer" for your AI quality. No ink required.
us02web.zoom.us/webinar/regi...
#AIQuality #SoftwareEngineering #Webinar
We replaced stack traces with a haystack made of other haystacks. When the output is "off," who's the suspect?
Prompt Prep? RAG retrieval? Context "Glue"? The Model?
Stop vibe-checking. Start engineering.
Read the blog:
testingil.com/2026/01/why-...
#AIQuality #SoftwareTesting #RAG
We replaced stack traces with a haystack made of other haystacks. When the output is "off," who's the suspect?
Prompt Prep? RAG retrieval? Context "Glue"? The Model?
Stop vibe-checking. Start engineering.
Read the blog:
testingil.com/2026/01/why-...
#AIQuality #SoftwareTesting #RAG
"The vibe is off" isn't a bug report. It's a cry for help.
Stop shipping AI based on a "Vibe Check." Join me Feb 18 for The Anatomy of AI Quality webinar. It's time to build a Radar for the fog and regain control of the wheel.
us02web.zoom.us/webinar/regi...
#AIQuality #SoftwareEngineering
Want to level up your team's testing? I train teams in unit testing, TDD, API, BDD, and Clean Code.
Now including AI in testing:
Using AI as a copilot
Testing AI products
Let's get your team out of the bug-fixing rut.
Ping me to cook up some quality.
#AIQuality #QA #QAStrategy #SoftwareEngineering
In the old days, a bug was a crash. Today, a bug can be a "correct" answer that costs 5x the budget to generate.
We need a new categorization: User, Ops, Hybrid.
Watch the video
youtu.be/f3nII_SsbCc
#AIQuality #SoftwareTesting #GenAI #QualityEngineering
In the old days, a bug was a crash. Today, a bug can be a "correct" answer that costs 5x the budget to generate.
We need a new categorization: User, Ops, Hybrid.
Watch the video:
youtu.be/f3nII_SsbCc
#AIQuality #SoftwareTesting #GenAI #QualityEngineering
Modern development ain't easy.
Old "pass-fail" don't work on non-deterministic code. It's like using leeches for modern diseases.
I help R&D leaders make sense of their AI apps quality.
Stop gambling. DM me.
#AIQuality #QA #QAStrategy #SoftwareEngineering
Modern development ain't easy.
Old "pass-fail" don't work on non-deterministic code. It's like using leeches for modern diseases.
I help R&D leaders make sense of their AI apps quality.
Stop gambling. DM me.
#AIQuality #QA #QAStrategy #SoftwareEngineering
Remember SQL Injection? Simple times.
Now we have Prompt Injection. The art of convincing your AI to ignore instructions.
From buying a car for $1 to pirate jokes - it sounds funny until it happens to you.
Start thinking like attackers.
youtu.be/vc-rJifDBM4
#PromptInjection #AIQuality
Modern development ain't easy.
Old "pass-fail" don't work on non-deterministic code. It's like using leeches for modern diseases.
I help R&D leaders make sense of their AI apps quality.
Stop gambling. DM me.
#AIQuality #QA #QAStrategy #SoftwareEngineering
LLMs’ impact on science: Booming publications, stagnating quality #Science #ComputerScience #ArtificialIntelligence #LLMImpact #SciencePublishing #AIQuality
LIVE IN 2 HOURS!
The production-ready methodology for AI Quality.
Drift. Evolution. Defense.
Grab your spot now:
us02web.zoom.us/webinar/regi...
#AITesting #Webinar #TechLeadership #AIQuality
