The GeekNarrator - Databases, Distributed systems, Stream Processing, Language internals. Software Engineer(Flink Engine) @Confluentinc. Opinions are my own.
Berlin, Germany
Joined April 2022
- Tweets3.5K
- Following478
- Followers8.2K
- Likes3.9K
Pinned Tweet
The JVM internals series is growing 👇
The JVM Internals - Basics (Episode 1)
youtu.be/Yg69qM0iZa4
JVM Internals: Heaps, Stacks, Program Counters and more
youtu.be/MCTTg8VPaBY
Peek into your Class file? JVM Internals
youtu.be/15sz5yvQ-Po
How is memory allocated on the heap?
youtu.be/qfcnB8kciCE
JVM Internals - Garbage Collection Types and Their Internals
youtu.be/p1Yamdly0QE
How do Garbage Collection Algorithms work?
youtu.be/LeYssTeCSFg
Are Garbage Collection Algorithms Complex?
youtu.be/1TH7NZisw3o
G1GC - Low Latency High Throughput Garbage Collector
youtu.be/SkZxGFnTK3Y
ZGC : The Java Garbage collector explained
youtu.be/hGrWF1Pk4Dk
Please join the membership to watch the full series and get access to other member-only playlists 🚀
The best way to handle a mistake is to completely accept and own it. No ifs and no buts.. no nuanced justification.. no playing with words….
A simple “we messed up and we will do better” works just fine.
in 2024 when I did a podcast on SlateDB with @criccomini I never imagined I would contribute to SlateDB...
exactly 2 years later (was also October) I joined SlateDB as a committer. 🥳
Such an honour. Such a friendly community. Excited to contribute further.
Excited to have @thegeeknarrator join SlateDB as our latest committer! He's been doing a bunch of work on splitting/merging DBs and also did some dataloss edge case fixes for us. :) Welcome!
Writing is Thinking.
If you feel you are losing your ability to think, you are not alone.
I feel something similar from time to time, if I don't write something on my own.
To avoid getting worse at it, try to be alone with a notebook or an editor with NO AI access and write on a topic that is new to you and a bit challenging for at least 30 mins a day.
I am sure you will start thinking again.
Top 5 bottlenecks:
- Human imagination
- Human attention bandwidth
- Human articulation
- Legacy processes made for humans, used for agents
- Legacy infrastructure made for humans, used for agents.
Massive 🚀 congratulations @apurva1618 @almoggavra
I’m excited to announce that Responsive has been acquired by Databricks to accelerate the development of Lakebase and Neon!
We couldn't be more excited to join the Lakebase team. We've long admired their technical vision, and the more we got to know them, the more impressed we became by their ambition, execution, and momentum. It feels like a natural home for us.
I want to thank Responsive’s customers, investors, and especially the team. More than anything, I’m grateful for the trust you’ve placed in us through the good times and the tough times.
Onward and upward!
Kaivalya Apte - The Geek Narrator retweeted
I could easily lose an afternoon clicking through these topics. That visual layout makes technical podcasts much easier to browse.
I work on Distributed systems so naturally my interests are all over the place 👋
Hey I am Kv/Kai - an engineer working on Distributed systems and Databases, and someone who finds himself asking "but how does it actually work?" a lot.
Asking that question has led me to create 150+ technical podcast episodes and tutorials. Kafka, database internals, the JVM, streaming systems, engineering and AI.. plenty of rabbit holes with great company of experts along the way.
I refactored my website to help you easily find the right rabbit hole for you. Something I chose to call "Knowledge Atlas"
Give me a follow and have a look at the website to find some great episodes on the channel aligning with your interest.
Feel free to comment and introduce yourself so I can follow you back!
Cheers,
Kv
I work on Distributed systems so naturally my interests are all over the place 👋
Hey I am Kv/Kai - an engineer working on Distributed systems and Databases, and someone who finds himself asking "but how does it actually work?" a lot.
Asking that question has led me to create 150+ technical podcast episodes and tutorials. Kafka, database internals, the JVM, streaming systems, engineering and AI.. plenty of rabbit holes with great company of experts along the way.
I refactored my website to help you easily find the right rabbit hole for you. Something I chose to call "Knowledge Atlas"
Give me a follow and have a look at the website to find some great episodes on the channel aligning with your interest.
Feel free to comment and introduce yourself so I can follow you back!
Cheers,
Kv
Kaivalya Apte - The Geek Narrator retweeted
This is now also available on my website as a blog.
Visuals and examples all from the video. Check it out!
geeknarrator.com/blog/skip-l…
In this world, where AI can explain you anything you need, I think a human touch is needed. So I created a video explaining how:
A linked-list + Coin Flips = expected O(log n) search.
Skip Lists are one of those ideas that make you stop and ask: "How does that actually work?"
Watch here: youtu.be/2JJIqJ31oSo
This is now also available on my website as a blog.
Visuals and examples all from the video. Check it out!
geeknarrator.com/blog/skip-l…
In this world, where AI can explain you anything you need, I think a human touch is needed. So I created a video explaining how:
A linked-list + Coin Flips = expected O(log n) search.
Skip Lists are one of those ideas that make you stop and ask: "How does that actually work?"
Watch here: youtu.be/2JJIqJ31oSo
In this world, where AI can explain you anything you need, I think a human touch is needed. So I created a video explaining how:
A linked-list + Coin Flips = expected O(log n) search.
Skip Lists are one of those ideas that make you stop and ask: "How does that actually work?"
Watch here: youtu.be/2JJIqJ31oSo
“Claude has written for me 400 tests, and I have no confidence whatsoever in any of these tests.”
Sound familiar?
@bugarela on AI-written tests, formal specs, and why confidence needs understanding.
Full conversation: youtu.be/j-j6aGy-FF8
My X feed be like..
Today we launched a new model….
A DB company acquired another DB company…
Too much going on to follow. 😴
Kaivalya Apte - The Geek Narrator retweeted
Also on YouTube: youtu.be/j-j6aGy-FF8
Imagine Griffin saying sh*t like:
- Let’s unpack this
- This isn’t just about X, its about Y
- One honest caveat, I haven’t confirmed any of this… 🤡
but as a real looking person
Introducing Griffin, the first model to pass the video Turing test.
48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video.
It’s the first Human Interaction Model (HIM).
Readers added context they thought people might want to know
The 48% figure and "video Turing test" claim are from Tavus's own study of 54 one-minute calls, not independently verified or using a standard protocol. Griffin-Lite leads NVIDIA's VideoFDB benchmark on their public leaderboard.
cellcog.ai/blog/tavus-gri…
research.nvidia.com/labs/amri/proj…
tech-ish.com/2026/10/02/tav…
Kaivalya Apte - The Geek Narrator retweeted
It's been a while since a was on a podcast, so lot's of new ideas on this one! So fun to talk about executable specs💜 Thanks @thegeeknarrator for hosting!
AI writes the code.
AI writes the tests.
Everything passes.
What have we actually proved?
I sat down with Gabriela Moreira @bugarela to explore what it takes to trust AI-written software: what tests miss, who checks an AI-generated specification, and where Quint @quint_lang and formal methods can help.
If you’re shipping AI-generated code, this conversation is for you.
Chapters:
00:00 Introduction
01:35 AI, correctness & the human in the loop
03:24 What does “correct” actually mean?
07:36 Who checks an AI-written specification?
13:43 Complexity, failures & the real world
19:26 What belongs in a model?
25:16 Why Quint exists
32:51 Can we trust our tests?
35:32 Keeping code & specifications in sync
41:14 Behaviours as paths through a state machine 47:07 Confidence through understanding
53:04 Quint in practice
58:53 Where to go next
Full episode below.
AI writes the code.
AI writes the tests.
Everything passes.
What have we actually proved?
I sat down with Gabriela Moreira @bugarela to explore what it takes to trust AI-written software: what tests miss, who checks an AI-generated specification, and where Quint @quint_lang and formal methods can help.
If you’re shipping AI-generated code, this conversation is for you.
Chapters:
00:00 Introduction
01:35 AI, correctness & the human in the loop
03:24 What does “correct” actually mean?
07:36 Who checks an AI-written specification?
13:43 Complexity, failures & the real world
19:26 What belongs in a model?
25:16 Why Quint exists
32:51 Can we trust our tests?
35:32 Keeping code & specifications in sync
41:14 Behaviours as paths through a state machine 47:07 Confidence through understanding
53:04 Quint in practice
58:53 Where to go next
Full episode below.