Brian Zhan (@brianzhan1) started following @UberEng on Oct 9, 2024

Discover how @Uber made experiment evaluations 100x faster! We’ve drastically reduced experiment evaluation latencies from milliseconds to microseconds by shifting from a remote to a local evaluation architecture. This improvement supercharges A/B testing across Uber's backend

Discover how Uber scales Presto for efficiency 🚀 Presto, Uber’s open-source SQL query engine, powers over 20 clusters and processes half a million queries daily. With 7,000 weekly users, Presto supports everything from analytics to data-driven services. Find out how we’re

RT @databricks: Excited to share that #DatabricksVentures has invested in @VoyageAI's Series A funding round! We will also soon offer Voy…

RT @mdredze: Thrilled to see @JohnsHopkins affiliated with 2 of the 4 🎖️outstanding🎖️ paper awards at #COLM2024 🥳 Dated Data @jeff_cheng_7…

RT @mrdrozdov: Those are some outstanding papers! Huge congrats to recent Databricks x Mosaic intern @ruyimarone and the stellar team of c…

Amazing that you can now train ACT on a macbook, with built-in camera for observation and a <$500 robot to run the policy. A large step in democratizing robot learning. Congrats @RemiCadene @huggingface!

Silicon Valley can keep convincing themselves they are the best at “software”, “AI”, until a Chinese hardware startup pulls off the best AI-driven locomotion on a production humanoid. New hardware companies from China won’t be satisfied with manufacturing, unlike their

All jobs submitted between 6am-11am PT failed due to a server outage from a third-party provider. The issue is now fixed–please re-submit failed jobs. We apologize for the inconvenience.

We deployed the following improvements to the Chai-1 lab today based on community feedback: - Support for PTMs, by using CCD codes - Email updates for completed jobs - Confidence scores (eg plddt) saved to PDB files - Faster computations

RT @ido_pesok: ⚡️🎊 Spark Email is here!! Demo below : ) Scheduling is a pain for many teams and individuals. Spark solves this painpoint.…

RT @OpenAI: Announcing GPT-4, a large multimodal model, with our best-ever results on capabilities and alignment: https://t.co/TwLFssyALF h…

RT @drdannywilliams: Late chunking - a paper I co-authored with @JinaAI_ - shows that switching the order in which you chunk and embed you…

Really impressed by this approach and how well this new model performs on benchmarks esp given its relatively small parameter count. More importantly, I love this trend of investing in new approaches and research for embedding models. LLM's have gotten a lot of love for the

RT @cognition_labs: We worked closely with OpenAI over the last few weeks to evaluate OpenAI o1's reasoning capabilities with Devin. We fou…

RT @tryramp: Honored to make our debut on the Forbes Cloud 100 list. This recognition, just years after our founding, reflects our team's…

RT @ecardenas300: Language models paired with function calling (or tool use) is a powerful way to build AI systems 🤖 Developers define a s…

RT @aiDotEngineer: AI recaps of @itsSandraKublik @cohere Command R talk thanks to @_thothai x @basedsocialco https://t.co/nnfX1M71uB

cc: @tianle_cai

RT @scaling01: i love the @dwarkesh_sp podcast my favorite episodes is probably one of these: Sholto & Trenton Patel & Asianometry Aschenb…

No matter what happens, this image is worth it.

RT @SonnyBunch: I’m sorry, this is the funniest single photo of the entire election cycle. https://t.co/w6PuJlPPmW

RT @CRV: Congratulations to @tengyuma on today’s Series A. @VoyageAI’s innovative approach uses vector embeddings trained on a company’s da…

We're excited to be joined by @SnowflakeDB in the Series A round for Voyage AI, and to offer Voyage AI embedding models for RAG to all users on Snowflake Cortex. Within months, Voyage sold their embedding models to over 250 customers, and this will turbocharge adoption

Over the past months, @VoyageAI has grown rapidly to over 250 customers. We're excited to move to the next stage and partner with @databricks to make the Voyage embedding models for RAG available broadly.

and experimentally they are equivalent. https://t.co/HAkeTQ7CKx

graph convolution is, to some extent, equivalent to Mixup (in fact, their mathematical expressions are already quite similar). https://t.co/WNEsnZYCEf

RT @AdinHaykin1: I wonder why that "golden age" didn't last https://t.co/zXnbKq5tY4

RT @EylonALevy: There are approximately 130 Muslims for every Jew in the world, and around 8 Muslims for every Jew in Australia. We know th…

RT @appenz: Congrats to @tengyuma and the amazing team at @VoyageAI. It is often underestimated how much of an effect embeddings can have,…

RT @TaherHassonjee: Always nice to see the good guys win. Congrats @tengyuma and team!

RT @cameron_pfiffer: I've used voyage a bit and it's really solid! Super happy to hear this

RT @jxnlco: When will Voyage allow me to embed a whole Pydantic object?

RT @ShunyuYao12: Congrats @tengyuma and co, and congrats to @CRV and co😃

RT @minorun365: 埋め込み＆リランクのVoyageがさらなる資金調達に成功とのこと。頑張れ〜！！ ちなまにAWSではSageMakerでマケプレから利用可能。Bedrockにも来たら、さらにRAGの強みになりそう

RT @jobergum: Great to see more investments into making more unstructured data useful. I’m personally hoping for commercial ColQwen2 endpoi…

RT @frankzliu: I joined @VoyageAI last week to work on applied ML. The team is cracked — excited to be a part of this journey! Oh, and fee…

RT @AlignsAI: This is super cool, congrats team @VoyageAI !!!

RT @1littlecoder: New player, New funding!

RT @mattlynley: A startup on my “watch list” from November 2023 🙂 https://t.co/KdCJMXHxSE

RT @shao__meng: # 祝贺 Voyage AI 完成融资 🎉 祝贺 Voyage AI @VoyageAI @tengyuma 完成新一轮融资, 由 @CRV 领投, 融资总额已达 $28M 💵 Voyage AI 专注于非结构化数据搜索和检索, 提供嵌入模型和…

Exciting to see the state-of-the-art in text embedding being built over Nomic open research! Open source results in massive impact

Explore the lives and legacies of over 1.4 million individuals from English Wikipedia’s biographical articles in our Atlas map here: https://t.co/Rq5cBLwxrn