Peter Ronning @peter_ron1

Senior Data Scientist | AI/ML Engineer | PhD Researcher @UW CEO of @further_ww San Francisco Joined March 2025

Tweets

46
Followers

56
Following

154
Likes

97

Further @further_ww

4 months ago

Further's Jason Tabeling breaks down how to leverage the power of BigQuery ML to optimize your targeting, bidding strategies, and ultimately, your ROI in his article on Search Engine Land. ow.ly/wzmB50VGtv6 #GoogleAds #BigQueryML #MarketingAnalytics #PPC

0 1 1 111 0

Alexander Novikov @SashaVNovikov

4 months ago

After 1.5 years of work, I'm so excited to announce AlphaEvolve – our new LLM + evolution agent! Learn more in the blog post: deepmind.google/discover/blog/… White paper PDF: storage.googleapis.com/deepmind-media… (1/2)

53 388 3K 360K 2K

Download Image

elvis @omarsar0

4 months ago

Reasoning LLMs Guide Here is my practical guide to building with Reasoning LLMs. Lots of dev tips in it. It covers: - What are Reasoning LLMs? - Top Reasoning Models - Reasoning Model Design Patterns & Use Cases - Reasoning LLM Usage Tips - Limitations with Reasoning Models

10 147 711 85K 1K

Download Image

elvis @omarsar0

4 months ago

WebThinker combines large reasoning models with deep research capability. This paper introduces a reasoning agent framework that equips large reasoning models (LRMs) with autonomous web exploration and report writing abilities to overcome limitations of static internal…

6 78 297 21K 237

Download Image

Nikita Bier @nikitabier

4 months ago

The entire tech community is under the impression that AI coding will result in power flowing from engineers to “idea guys.” Wrong—it will always flow to whatever still has scarcity: those who know how to get distribution

255 389 6K 494K 2K

Percy Liang @percyliang

4 months ago

Announcing VHELM v2.1.2 for VLMs: We added the latest Gemini models, Qwen2.5-VL Instruct models, GPT 4.5 preview, o3, o4-mini, and Llama 4 Scout/Maverick. Prompts and predictions can be found on our website: crfm.stanford.edu/helm/vhelm/v2.…

3 26 70 22K 31

Download Image

Peter Ronning @peter_ron1

4 months ago

I think $GORK can be largely improved

0 0 0 349 0

Rulin Shao @RulinShao

4 months ago

Meet ReasonIR-8B✨the first retriever specifically trained for reasoning tasks! Our challenging synthetic training data unlocks SOTA scores on reasoning IR and RAG benchmarks. ReasonIR-8B ranks 1st on BRIGHT and outperforms search engine and retriever baselines on MMLU and GPQA🔥

5 64 354 51K 179

Download Image

Andrew Lampinen @AndrewLampinen

4 months ago

How do language models generalize from information they learn in-context vs. via finetuning? We show that in-context learning can generalize more flexibly, illustrating key differences in the inductive biases of these modes of learning — and ways to improve finetuning. Thread: 1/

8 152 768 98K 689

Download Image

elvis @omarsar0

4 months ago

Noticing a lot of bad AI product experiences these days. Lots of products feel rushed and unpolished. It's because AI is in its infancy, but already quite capable. There is a lot to unlock and improve (e.g., better UI/UX). Insane amount of opportunities to build and disrupt!

9 8 79 12K 22

Polina Kirichenko @polkirichenko

4 months ago

We are hiring a PhD research intern at FAIR w/ @marksibrahim @kamalikac to start this summer or Fall! Potential topics: trustworthy and reliable LLMs, multi-modal LLMs and agents, post-training, reasoning, with a focus on open science/sharing our findings in the paper at the end…

5 36 403 49K 242

Nikita Bier @nikitabier

4 months ago

There are two ways to validate a network-based product: 1. The best method is repeatedly running experiments into small, closed communities because you get so many shots at bat to get the product right. However, it only works for communities that are: • Plentiful (i.e.,…

68 55 1K 190K 1K

Delip Rao e/σ @deliprao

4 months ago

silly comments are why code generation models can reason better and write better code. Stripping off silly comments post hoc is a more straightforward task, with today’s LLMs, than building accurate codegen models that are not verbosely commenting.

Ofir Press @OfirPress

4 months ago

12 0 79 13K 6

Download Image

3 4 31 4K 3

Nikita Bier @nikitabier

4 months ago

Wow that’s a dumb idea. Someone already made it before—with the worst user experience possible and with the least memorable name a human could think of. If theirs didn’t work, there’s no way yours is going to work.

155 145 4K 361K 919

Peter Ronning @peter_ron1

4 months ago

You're spot-on, bots are evolving but on X you catch those subtle tells, phrasing that's just too polished / slightly out of context.

Lee Robinson @leerob

4 months ago

You're spot-on, bots are evolving but on X you catch those subtle tells, phrasing that's just too polished / slightly out of context.

64 20 1K 170K 160

1 0 1 236 0

Julien Chaumond @julien_c

4 months ago

BOOOOM: Today I'm dropping TINY AGENTS the 50 lines of code Agent in Javascript 🔥 I spent the last few weeks working on this, so I hope you will like it. I've been diving into MCP (Model Context Protocol) to understand what the hype was all about. It is fairly simple, but…