Navarre Trousselot

speaker

492 appearances 1 recordings 1 series first heard Jun 2026 last heard 10 Jun

Navarre Trousselot’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 1 in all, peaking in Jun 2026 with 1.

Appearances

newest first · ▶ plays the moment

Australian Finance Podcast · ETF tax traps, CGT and cost base adjustments with Navarre Trousselot · 10 Jun 2026 podcast

But the cool thing is with Nevexa AI and MCP server, which is just a fancy integration to the Claude and ChatGPT, you will get the benefit of being able to use Nevexa's code and logic and...

memory and all the settings being stored correctly, but then be able to use the power of AI to kind of do some interesting things with it.

And so we definitely don't want to, we don't want to be one of those businesses that like, you know, closes off all the data and you have to use Nevex or AI and you can't like

export it we want to be an open data platform so we know how to store this type of data in the best way ai can do stuff we'll never be able to do um which because it's just so crazy and hardcore so we want to unlock those benefits still for our customer base one thing important to note though is um the way ai models generate results and um

And this is a great question because most people have no idea because they assume, you know, it's a computer.

Computers are good at maths, right?

So it should be perfect.

But

Large language models aren't doing maths at all.

They see your equation of one plus one.

And usually the next character that follows is a two.

And that's how it's doing its arithmetic.

Yeah, exactly.

So that's how you can fool it pretty easily into hallucinating maths.

And you can get it to like count the number of letters in a particular word and it can stuff that up because in its wisdom it can't figure some of those things out.

001101.

And that's the problem is because you... And even myself, I expected it was doing maths because it's a computer, right?

And it is...

depending on the models you use, now it can, behind the scenes, it will actually create code to do the maths to then return you the result.

But it happens different every time.

Showing 461–480 of 492 · page 24 of 25