Ben Hamner

Sumble co-founder and CTO. Learning high-quality, structured data about the world. Formerly @kaggle

[Ben Hamner @benhamner]

Wordle xxx 1/6

🟩🟩🟩🟩🟩

Get Wordle right on your first guess using the daily ⬛🟨🟩 tweet distribution https://t.co/DolY2g0Xrj

[Ben Hamner @benhamner]

@bernhardsson @j_schottenstein @generalcatalyst @Redpoint Congrats!

[Ben Hamner @benhamner]

@tunguz I’ve spent a lot more time, not less, writing code with the latest generation of coding agents

[Ben Hamner @benhamner]

@jasonlk What volume of contacts? (total, and then the # of C-suites / champions that you especially care about having up-to-date)

[Ben Hamner @benhamner]

The “AI escape the box” experiment was comically naive in hindsight. There is no box. Instead, there’ll be embedded access to every critical system in the world

[Ben Hamner @benhamner]

What will engineering productivity look like in a year once a high fraction of codebases are written by Claude Code/Codex and no human has a detailed understanding of these parts anymore?

[Ben Hamner @benhamner]

if country==“AU”:
await asyncio.sleep(0.52)

[Shreya @Oblivious9021]

Interviewer:
Your page loads in 80 ms in Australia but 600 ms in India. Same backend. Same code. What would you use to fix this?

[Anthony Goldbloom @antgoldbloom]

Excited to share more publicly what @benhamner and I have been working on:
https://t.co/QXvodhsM5e

We are building a knowledge graph for GTM data & have early customers such as @elastic, @figma and @Snowflake

[Ben Hamner @benhamner]

english in 2030 : python :: python in 2020 : assembly

[Ben Hamner @benhamner]

WOW - cold inbound hiring processes are the worst I've ever seen it:

- Huge volume of applications (probably triggered automatically)
- Large volume of fake resumes
- Extensive use of AI assistants in phone screens

[Ben Hamner @benhamner]

@MeganRisdal eke, hope you fully recover quickly

[Ben Hamner @benhamner]

@Xfinity @XfinitySupport -

1. Internet has been down whole day in Palo Alto
2. Your outage map for Palo Alto is also not working ("address not found")
3. Your support page doesn't actually provide support

[Ben Hamner @benhamner]

@yanda @merci congrats!

[Ben Hamner @benhamner]

You could release a crazy stat on the number of times (and number of employees) ClosedAI companies have downloaded open models from HuggingFace

[clem 🤗 @ClementDelangue]

All closed-source frontier labs use tons of open-source all over the stack, starting from python, @PyTorch, @huggingface all the way down to RoPE, GQA, flash-attention, or any tiny improvements released by open-source players.

The whole transformers architecture (the T in gpT) comes from an open research paper and open-source model from 2017: https://t.co/01WrQqeKiU

This is ok but would be much nicer if they would acknowledge the contributions & contribute back more!

[Ben Hamner @benhamner]

@jasonlk dumb question: what’s the difference between the AI SDR and the AI BDR?

[Ben Hamner @benhamner]

@Nate_Keating Yeah, expect that number to go up orders of magnitude as video processing / generation + deep research / reasoning usage + context lengths explode

[Ben Hamner @benhamner]

This is around 5x the ~200 trillion natural language tokens humans generate every month

[Logan Kilpatrick @OfficialLoganK]

Google is processing 980 trillion+ monthly tokens across our products and APIs (up from 480T in May) 🤯

No slowdown in sight, intelligence is everywhere.

[Ben Hamner @benhamner]

Assumptions: 8 billion people * 20k spoken/written words per day * 1.3 tokens per word

[Anthony Goldbloom @antgoldbloom]

Kaggle launched an LLM eval product https://t.co/lJ8uSI2HmE

This has the potential solve the biggest challenge in the LLM ecosystem: strong and diverse evals

[Ben Hamner @benhamner]

@arrington Could always be for training / ongoing evaluation and quality control