Aziza Renner

Curious about how things work | Data & AI
Currently @justeattakeaway
Will interrupt an AI conversation to compliment your outfit

[Aziza Renner @zizuinbeta]

hey, i'm Zizu! just a girl who accidentally ended up in tech :)

studied law, worked in international organizations, took a detour through finance, and somehow landed in data & AI.

i like solving problems, asking too many questions, and occasionally pretending my side projects are under control.

here to share what i'm building, learning, and getting wrong along the way.

still in beta, obviously.

[Aziza Renner @zizuinbeta]

5/ The lesson I keep coming back to: each step should answer one question. Does it work? Will people use it? Only then, scale.

Longer write-up if you're working through the same thing: https://t.co/XWorhWoLHU?

[Aziza Renner @zizuinbeta]

1/ When we started building an AI tool to help account managers prep for restaurant meetings, the obvious first step looked like connecting all the data systems.

We didn't. A few things I learned from that:

[Aziza Renner @zizuinbeta]

4/ Then a request nobody expected: help planning driving routes. Faster prep meant more visits, and travel became the new bottleneck.

[Aziza Renner @zizuinbeta]

Hi X!
I’m Aziza 👋 I’m a data analyst and moved to Amsterdam in March!

Some facts about me:
-Lived in Italy for 4 years 🇮🇹
-Have both a bachelor’s and a master’s in law
-Took a turn from law into data, and now keep finding new rabbit holes in AI
-Especially curious about how AI remembers things and how we learn
-Also love to yap about fashion and a little bit of everything 😂

I’m still finding my people in Amsterdam and would love to meet others who enjoy exploring ideas, building things, or figuring out their next chapter!

I’ll be sharing what I’m learning and experimenting with in AI here and at https://t.co/8SmyqaiUnD.

Let’s connect — DM me, or let’s grab a coffee! ☕

[Aziza Renner @zizuinbeta]

https://t.co/VXS018QvKI

[Aziza Renner @zizuinbeta]

I’ve been going down a rabbit hole with AI memory lately, so I wrote about it on Substack.

How should agents remember a company’s decisions and share context? I tried a small experiment to understand it better.

Still lots to figure out but link to part one is down below :)

[Aziza Renner @zizuinbeta]

In every case, the person kept the judgment. The agent took the grind. Full write-up: https://t.co/SU2Nt66fE0

[Aziza Renner @zizuinbeta]

Unpopular opinion: AI agents aren't coming for your job.

They're coming for the worst part of it.

3 real examples I saw at work this year

[Aziza Renner @zizuinbeta]

Next, I’m planning to finish 3 more books and see what patterns emerge.

It's already making me more curious about what I read. Whether it actually helps my memory is still an open question.

If you work with language, memory, or knowledge graphs, how would you test this?

https://t.co/qKArqaQJpH

[Aziza Renner @zizuinbeta]

I can look up an English word, understand the sentence, and forget the word before the chapter ends.

Excellent short-term customer service from my brain.

I wanted to understand why this keeps happening. Somehow this now involves PostgreSQL.

[Aziza Renner @zizuinbeta]

Underneath: Next.js, FastAPI, PostgreSQL, OpenAI for English–Russian explanations, and Sigma.js for the graph.

It’s still running locally on my Mac. I’m building the tool as I use it, which is a very effective way to discover that my original design made no sense.