AIEdTalks's profile picture. 18+ yrs in AI | 35+ patents | 17+ papers | Helping professionals & orgs grow with AI, career shifts & mindset. Tweets = clarity, not chaos

AI EdTalks

@AIEdTalks

18+ yrs in AI | 35+ patents | 17+ papers | Helping professionals & orgs grow with AI, career shifts & mindset. Tweets = clarity, not chaos

Pinned

A collection of threads on AI Agents-Fundamental, research problems, solutions, git repos, and a lot more. Subscribe to my newsletter for detailed learning (link in bio). Like, retweet, and follow for new learning every day!


5 Signs Your Agentic System is struggling with Context Problems: - It forgets earlier instructions - It contradicts itself mid-task - It overuses irrelevant data - It fails to adapt tone or role - It slows down as context grows To fix the above issues, look into your context…


Laying off because of AI is a lie ! Question: What are the real reasons behind these layoffs?

There have been significantly more layoffs in 2025 that companies attributed to AI than what’s being officially reported. Is it possible that companies are being dishonest and using AI as an excuse? 🤔

MauiBoyMacro's tweet image. There have been significantly more layoffs in 2025 that companies attributed to AI than what’s being officially reported.

Is it possible that companies are being dishonest and using AI as an excuse?   🤔


AI EdTalks reposted

I have been working on AI agents for the last year. I have built many agents, worked on various research problems, and always had many questions in my mind: - Converting existing APIs to a tool and handing it to an Agent is enough? - When to build an agent does not make…


AI EdTalks reposted

Should we allow engineers to use coding agents during interviews? Engineer leaders/Hiring managers- What are your thoughts?


AI EdTalks reposted

Your AI agents are failing. But not because of the model. They're failing because of context. When context is incomplete, agents hallucinate. When it's ambiguous, they make wrong decisions. When it's outdated, they fail silently. Context quality determines agent reliability…

AIEdTalks's tweet image. Your AI agents are failing. 
But not because of the model. 
They're failing because of context.

When context is incomplete, agents hallucinate. 
When it's ambiguous, they make wrong decisions. 
When it's outdated, they fail silently.

Context quality determines agent reliability…


IndQA- Indian language benchmark!

Introducing IndQA — a new benchmark that evaluates how well AI systems understand Indian languages and everyday cultural context. openai.com/index/introduc…



I also believe it's not AI!

Optics y'all! IBM is doing layoffs today. AI causing layoffs is a lie. Remember - in 2023 they cut 8,000 people in HR (mostly) while rolling out AskHR. Not “AI causing layoffs”… AI used as cover to cut people. Now they’re waving a $150B “investment in America” flag… and…

thejobchick's tweet image. Optics y'all!
IBM is doing layoffs today. 

AI causing layoffs is a lie. 

Remember - in 2023 they cut 8,000 people in HR (mostly) while rolling out AskHR.

Not “AI causing layoffs”… AI used as cover to cut people.

Now they’re waving a $150B “investment in America” flag…
and…


Another company, same story. Feel bad that those are getting hit.

IBM CUTTING 8,000 JOBS IN LATEST LAYOFFS



AI EdTalks reposted

Your AI agents are failing. But not because of the model. They're failing because of context. When context is incomplete, agents hallucinate. When it's ambiguous, they make wrong decisions. When it's outdated, they fail silently. Context quality determines agent reliability…

AIEdTalks's tweet image. Your AI agents are failing. 
But not because of the model. 
They're failing because of context.

When context is incomplete, agents hallucinate. 
When it's ambiguous, they make wrong decisions. 
When it's outdated, they fail silently.

Context quality determines agent reliability…

United States Trends

Loading...

Something went wrong.


Something went wrong.