It was quick and this was my prompt: A memorable repetitive beat that can be danced to that sounds like house music with a rap beat. suitable for studying. Instrumental only.

#Gemini #GoogleAI #Multimodal #AudioAnalysis #AI #BerniePng #TechTrends #Productivity #LLM #Gemini15Pro #AITesting #FutureTech #VoiceAI #MachineLearning #Innovation

March has been crazy. Everyday I'm bombarded by AI news – launching of new LLMs, new Claude features, a better way to do things, combining this to that, datacenters sprouting up here and there – and even for someone who loves change, the velocity of AI changes is honestly hard to keep up.

I'm (still) loving every bit of it, and my brain is on fire. Check in on me in 2 months, I might have a different story to tell. Might be fried by then.

Three key mini inflections took place in my life with AI this month:

• 1st week of March

  • The Deliveroo exit from Singapore. I did a symbolic mini tour around almost all the zones, March 1 – 4. It was nice.
  • My attempt to get hired as an apprentice with AI Singapore. On March 4 they sent out the technical assessment to everyone who applied for the apprenticeship program, giving us 6 days to work on it then submit. Wish me luck.

• 2nd week of March

  • I met up with a few of my ex-colleagues. We talked about the future and our plans. It was nice to catch up with them.

• 3rd week of March

  • With the GTC 2026 and SXSW keynotes combines, I finally understood where we are heading, and really gave me a lot to think about, mainly how what value can I offer in this new world, and what gaps I have.
  • I paid for Claude for a month, but have not explored Cowork or Code in depth. I'm still able to build lots of stuff just using the chat.
  • My class assignment for this module 2 involves Big Data. I ingested 23 million rows from a public dataset and did analysis on it. It was fun.

At some point in the keynote Jensen said that in the future we would not just be paid in money, aka salary, but also, with tokens. I believe him. But what does it mean for us, the common folk? He talked about AI factories and showing off all his highly efficient racks with awesome “tokens/watt” servers, and how everything will be opensource. It's like “here's all the free chocolate and candy you can eat, but you have to pay for the teeth.”

Welcome to the new world.

#AI #LLM #FutureofWork #JensenHuang #OpEx #TechnologyTrends

What would a truly localized Singaporean LLM look like? Beyond just Singlish, I explore how a 'fellow citizen' AI could navigate our unique cultural nuances.

---

As I'm going through the free Stanford Online course CME295, I thought about Natural Language Processing (NLP) and how we currently tokenize English text to feed into LLMs. If we transform words into numbers, how do we add in culture nuances, slangs, etc that are unique to a country so that we can create a sort of naturalized citizen? What is Singapore currently doing?

So I asked Gemini for some thoughts: so if I am a citizen of a country who wants to help build a realistic LLM for my country where we have 4 languages and we want to not only create a balanced LLM, but also to incorporate our country's culture and slangs into it so that the LLM performs like a "fellow citizen" and knows all our nuances and history. How should I start?

Gemini's answer: “This is a noble and technically fascinating mission. Building a “Sovereign LLM” is the move from a model that knows about your country to a model that is of your country.

To make an AI feel like a “fellow citizen,” you have to move beyond just scraping the web. You need to follow a four-stage blueprint:

• Collection: Gathering multicultural datasets.
• Curation: Cleaning data to avoid “Data Poisoning.”
• Alignment: Teaching the AI nuances and cultural slangs.
• Evaluation: Testing if the AI actually sounds like a “Fellow Citizen.”

...

Be careful with “Data Poisoning.” If you only scrape one political side or one ethnic group's slang, your “Fellow Citizen” AI might end up being biased or exclusionary. Balance is your hardest job.”

It then went on to suggest some open-source models to start with and actionable next steps.

[NOTE] I'm using this blog article to document the start of my journey. It will be a multi-year quest of mine. I know Singapore is already embarking on building her own LLM or LLMs, but wouldn't it be great if I can create my very own Singaporean friend, a super knowledgeable, kind and caring one at that.

Cool idea?

#ai #LLM #NLP #SovereignAI #Singapore #Singlish #AISingapore #CME295 #LearningInPublic