pnote

Deep research at Off Avignon

🎭 3rd annual exercise in screening a festival program that's 1700 shows thick

Jul 27 • 

Przemek

4

February 2025

12 years at Google

📝 The golden age of Google as I’ll remember it; and what may lie beyond it

Feb 9 • 

Przemek

5

1

December 2024

Hash Code: AI from DeepMind beats top human programmers

🏆 A "Deep Blue" moment for competitive programming?

Dec 8, 2024 • 

Przemek

2

Fake teachers out there

🎞️ A YouTube lecture with a catch

Dec 1, 2024 • 

Przemek

4

6

November 2024

Working on my spoken French with voice recognition AI

🎙Give me a feedback loop and I'll dial down my Eastern-european accent

Nov 24, 2024 • 

Przemek

5

4

How to give feedback

🎤 A 2-minute masterclass in feedback at a Parisian comedy club

Nov 3, 2024 • 

Przemek

8

4

October 2024

This is not reasoning

LLMs caught cheating at a math test

Oct 27, 2024 • 

Przemek

3

Rest in pieces: end-to-end vs modular deep learning

🚗 How do you program self-driving cars, end-to-end vs modular deep learning, neural networks are software systems

Oct 20, 2024 • 

Przemek

4

Don't skip weeks

📝 "Never miss twice" is bad advice; "Given the smallest excuse, one will not work at all"; not feeding the gremlins.

Oct 13, 2024 • 

Przemek

5

6

Feeling overwhelmed

💫 My feeling of overwhelm is not about having too many things to do. It's about not seeing a clear way out of the chaos.

Oct 6, 2024 • 

Przemek

6

2

September 2024

Prompt injection is not about segmentation

🚧 Delimiters don't save us from pirate jokes; the separation of instructions and data doesn’t exist; LLMs are hard to steer

Sep 8, 2024 • 

Przemek

2

August 2024

The distractible LLM: prompt injection explained

🏴‍☠️ We ask to review an email draft and hear a pirate joke instead; LLMs are distractible; a trusted prompt with untrusted data equals trouble

Aug 25, 2024 • 

Przemek

2

3