Picture

Julia Neagu @JuliaANeagu

CEO & Co-Founder @QuotientAI ✨ formerly @GitHub @GitHubCopilot 🤖 reformed physicist 👩‍🔬 ~ opinions are my own ~ quotientai.co/post/hello-wor… Boston, MA Joined March 2013
  • Tweets
    801
  • Followers
    694
  • Following
    974
  • Likes
    951
Julia Neagu retweeted
qdrant_engine Profile Picture

Qdrant @qdrant_engine

3 days ago

Feeling like your RAG model is just... okay? Let’s change that! 💡 Step beyond the ordinary and join us on May 7th for an exclusive webinar hosted in collaboration with @QuotientAI. Register in the link bellow 👇 buff.ly/4doMlwn

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

5 days ago

Are models overfit on public benchmarks or are public benchmarks so generic and representative of the general training corpus that regular training looks like overfitting 🤔 “overfitting” implies some intent/error during training, which may not be the case.

alexandr_wang Profile Picture

Alexandr Wang @alexandr_wang

7 days ago

Are models overfit on public benchmarks or are public benchmarks so generic and representative of the general training corpus that regular training looks like overfitting 🤔 “overfitting” implies some intent/error during training, which may not be the case.

Julia Neagu retweeted
QuotientAI Profile Picture

Quotient AI @QuotientAI

6 days ago

💡 Prompt optimization reveals different dominant models for domain-specific tasks. @Meta Llama3 performs best w/ generic prompt but @OpenAI takes the lead with more tailored prompt. 👆 This means that improving quality is an iterative process not a one-time model choice!

Julia Neagu retweeted
QuotientAI Profile Picture

Quotient AI @QuotientAI

6 days ago

We benchmarked @Meta Llama3 @databricks DBRX @MistralAI Mixtral-8x22b and @OpenAI GPT4 on our Product Catalog Q&A dataset! 1⃣ Llama-3-70b matches GPT4's pace! 2⃣ Llama-3-8b and Mixtral-8x22b have almost identical performance. 3⃣ DBRX is definitely the chattiest of the bunch 😉

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

a week ago

We've partnered with @qdrant_engine to help developers build the best AI products for their users! 🚀

QuotientAI Profile Picture

Quotient AI @QuotientAI

a week ago

We've partnered with @qdrant_engine to help developers build the best AI products for their users! 🚀

Julia Neagu retweeted
QuotientAI Profile Picture

Quotient AI @QuotientAI

2 weeks ago

Full house at @TechCrunch Early Stage yesterday for a panel on TAM in new and existing markets - featuring @tecoker @nabsicle @alex and our CTO @freddie_v4.

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

2 weeks ago

Congrats on the awesome panel @tecoker (@felicis), @nabsicle (@CellinoBio), @freddie_v4 (@QuotientAI) and @alex (@TechCrunch)!

tecoker Profile Picture

Tobi Coker @tecoker

2 weeks ago

Congrats on the awesome panel @tecoker (@felicis), @nabsicle (@CellinoBio), @freddie_v4 (@QuotientAI) and @alex (@TechCrunch)!

Julia Neagu retweeted
JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

3 weeks ago

@kevinroose There are two separate eval challenges that often get conflated: #1 LLM providers benchmarking their models and #2 AI companies evaluating LLM systems. #1 needs the same out-of-sample yard stick to show relative improvements, #2 needs domain specificity and customization.

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

4 weeks ago

This is a very important point that most people don’t realize — if you have a multi-step process with an error rate at each step, those error rates compound. The solution is to either improve the reliability or catch and correct the errors on the go — ideally both.

random_walker Profile Picture

Arvind Narayanan @random_walker

4 weeks ago

This is a very important point that most people don’t realize — if you have a multi-step process with an error rate at each step, those error rates compound. The solution is to either improve the reliability or catch and correct the errors on the go — ideally both.

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

4 weeks ago

Super excited about this one -- the AI space desperately needs frameworks to tell apart facts and fiction. Congrats @random_walker and @sayashk 👏👏👏

sayashk Profile Picture

Sayash Kapoor @sayashk

4 weeks ago

Super excited about this one -- the AI space desperately needs frameworks to tell apart facts and fiction. Congrats @random_walker and @sayashk 👏👏👏

Julia Neagu retweeted
freddie_v4 Profile Picture

Freddie Vargus @freddie_v4

4 weeks ago

Hello world 👋 excited to announce what @JuliaANeagu and I have been working on -- @QuotientAI Quotient enables developers to evaluate, improve and ship high quality AI products through fast, real-world, data-backed experimentation. 1/3

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

a month ago

That’s because most implementations of LLM-as-a-judge don’t make sense for what they’re supposed to evaluate. Ideally you can always use a mix of human, heuristic, and LLM eval. There are also tasks that are too unstructured for heuristics but structured enough for good LLM-eval.

ajratner Profile Picture

Alex Ratner @ajratner

a month ago

That’s because most implementations of LLM-as-a-judge don’t make sense for what they’re supposed to evaluate. Ideally you can always use a mix of human, heuristic, and LLM eval. There are also tasks that are too unstructured for heuristics but structured enough for good LLM-eval.

JuliaANeagu Profile Picture

Julia Neagu @JuliaANeagu

a month ago

AI progress is starting to bottleneck on good data before it bottlenecks on compute. @freddie_v4 built an interface that enabled a global team of volunteers create the datasets behind @CohereForAI’s Aya.

CohereForAI Profile Picture

Cohere For AI @CohereForAI

a month ago

AI progress is starting to bottleneck on good data before it bottlenecks on compute. @freddie_v4 built an interface that enabled a global team of volunteers create the datasets behind @CohereForAI’s Aya.

Trends for United States

Madrid
1,57 Mn posts
Joselu
317 B posts
Neuer
166 B posts
Harry Kane
77,1 B posts
Shea Langeliers
Steve Albini
60,2 B posts
Spring Hill
1.187 posts
Tuchel
132 B posts
Cable
23,3 B posts
Davies
83 B posts
#tnwx
3.942 posts
Raskin
22,4 B posts
TORNADO EMERGENCY
3.667 posts
Race Hub
#XMen97
67,7 B posts
Paul Skenes
6.017 posts
#ChampionLeague
12,7 B posts
#APorLa15
149 B posts
Marjorie Taylor Greene
26,1 B posts
Vini
139 B posts

You might like

DAAD Scholar @ Stellenbosch University|| Mathemathical Ecology (Elephant Conservation) || R || Public Speaking || Gospel
Victoria Okeowo
@VictoriaOkeowo
151 Followers
Full-service accounting + CFO firm for startups. We're here while you build your dream (and for everything in between). Led by @marcelisremco & @michaelbudnow
Standard Ledger
@standardledger
511 Followers
CTO & Co-founder @quotientai Research @cohereforai — past: @github Copilot, @quantopian — Tico 🇨🇷🇺🇸
Freddie Vargus
@freddie_v4
632 Followers
Trying to reduce suffering and make the future better
Josh You
@justjoshinyou13
1K Followers
visionary behind 3am release of 3am edition of tswift’s ‘midnights.’ @replicate @retool @github. bay area. via pdx, sf, chicago, and the ozarks.
Cory Wilkerson
@corywilkerson
2K Followers
working on @GitHub Actions. he/him. https://t.co/51WFDzUONf
Josh Gross
@joshmgross
132 Followers
Gamedev. PAST: Chrome Death https://t.co/RHY11u4zbI CURRENT: Den vänstra handens stig https://t.co/j7ajUPYomW FUTURE: https://t.co/srSvGXdRSf
M.James Short
@mjamesshort
830 Followers
Member of Technical Staff @Cohere
Wei-Yin Ko
@weiyinko_ml
120 Followers
Staff eng at @github. (She/her) Avatar by @deniseyu21, background by @nickwarner. DA-1482-1207-0163 🌴 cheshire137@ruby.social https://t.co/knGcy7ty6G
Sarah Vessels
@cheshire137
2K Followers
Ehpad des Jardins du Castel
Jardins du Castel
@JardinsCastel
551 Followers
Four letters | Business Analysis Guy | Husband | Small white ball chaser.
Ruf
@RufadzoB
533 Followers
Founder @Twain_ai - AI writing co-pilot for sales.
Mohamed Chahin
@mo_chahin1
294 Followers