Nick Omeyer
@NickOmeyer
Followers
153
Following
557
Media
13
Statuses
723
AI Engineering @ClickUp • Mostly lurking 👀
London
Joined April 2012
Jeff is super insightful & I enjoy his podcast appearances where hosts get him to unpack his ideas for plebs like me. But his writing often feels out of reach. So I unpacked Jeff’s Theory of MicroStrategy with the help of ChatGPT — great stuff 👌 https://t.co/EK72hN20pa
As $MSTR officially enters the NDX 100 today, here is my 1-page executive summary on the "Theory of MicroStrategy" and how to profit off of the unstoppable hyperfinancialization of finance. The trade of the decade is just getting started. Buckle up.
0
0
0
I went through the most popular AI repos on GitHub, categorized them, and studied their growth trajectories. Here are some of the learnings: 1. There are 845 generative AI repos with at least 500 stars on GitHub. They are built with contributions from over 20,000 developers,
44
298
1K
Fantastic series of posts! Full of tricks to get RAG systems working well. Hope you keep the series going @hrishioa 😇
Everything I'll forget about RAG part 3 is about something that bugs me in most RAG pipelines - the complete neglect of structure. https://t.co/pwHkqTIbS4 GPTs and HF assistants are pretty underused as didactic tools - so much easier to show rather than tell
1
1
3
Great tips particularly around storage & retrieval, thanks @sisilmehta 🙏
Practical Tips and Tricks used in a Production RAG Application There’s hundreds of RAG techniques, but the most useful ones are those that power a live LLM application. @sisilmehta from @heyjasperai presents some new best practices that helped his team deploy a production app
0
0
0
Nice overview of current approaches to building AI apps 👌
Interesting trend in AI: the best results are increasingly obtained by compound systems, not monolithic models. AlphaCode, ChatGPT+, Gemini are examples. In this post, we discuss why this is and emerging research on designing & optimizing such systems. https://t.co/tfnNuoTNNY
0
0
0
Chain-of-thought reasoning without prompting New paper by DeepMind suggests that CoT reasoning could emerge by changing the decoding process. Replace greedy decoding by investigating top-k token paths and find step-by-step thinking along some of them. https://t.co/iPrrJVPynh
0
0
0
Meta’s LLM for software testing work is super exciting. This paper describes Meta’s TestGen-LLM tool, which uses LLMs to automatically improve existing human-written tests. TestGen-LLM verifies that its generated test classes successfully clear a set of filters that assure
15
257
1K
Super exciting release 🔥 ColBERT is clearly the way to go and this makes it much more accessible to deploy in prod – thanks for the great work @jobergum & @vespaengine 🙏
Announcing ColBERT in @vespaengine, enjoy! - A new native Vespa ColBERT v2 embedder - ColBERT token-level vector compression (32x) - Support for long context via Vespa mixed tensors - Offload to disk - Eval Plus, it boasts the largest FAQ ever!😅 https://t.co/kNHMvXpV3e
1
1
8
New post: Will scaling work? This is the crux in arguments about AI timelines. In order to think through my own position, I wrote the post as a debate between a skeptic and a believer. Skeptic point 1: Data bottlenecks won't be clearer by self-play/synthetic data:
30
98
777
.@withmartian is very intriguing! @shriyashku describes the approach behind their model router here: https://t.co/PkpRP9HlYj I didn’t realise it was possible to inverse-distill models 🤯 Untangling the roles of neurons by mapping into a higher dimensional space, very cool 👌
Martian Raised $9M from @NEA @Prosus_Ventures and @GC and is launching its beta today! Our LLM router dynamically routes each request to the best LLM. We beat GPT-4 on @OpenAI's own evals (colab to replicate in thread) while cutting costs up to 98%+ https://t.co/JeSxnmU6sR 🧵
0
0
2
Interest in GenAI based on search volume Curious to see the US quite far “behind” and a large interest in audio gen from South America 🔊 Wonder what biases might be induced by the methodology 🤔
Pretty wild to see the largest adoption of generative AI coming from emerging countries like the Philippines > adjusted for population sizes See the interest by modality: - text - image - audio - video 🧵
0
0
0
Overview of the 4 top papers from #NeurIPS23 by @sophiamyang, 5 minutes per paper ⏱️🙏
📹 Deep dive into 4 NeurIPS 2023 best paper award winners: https://t.co/zfc0awC1IE - Are Emergent Abilities of Large Language Models a Mirage? https://t.co/WB10lJ1yYf - Scaling Data-Constrained Language Models. https://t.co/OMeehH8AMP - Direct Preference Optimization: Your
0
0
0
Detailed walkthrough on web scraping with GPT-4 from @itstimconnors — don’t brute force, look for specific keywords first
I spent the last 4 weeks building the holy grail of web scraping: It's a universal web scraper, built with GPT-4 + Vision + Headless Chrome The simple methods I tried at the start didn't work well: full page screenshots or the full HTML ended up being too much data to parse
0
0
2
If you're like me and couldn't make it to @NeurIPSConf this year, I made a reading list of papers that seemed relevant in the LLM setting! * Congrats to all papers at NeurIPS (best papers, orals, posters) * I made this list from a brief skim of best papers + orals, and mostly
0
0
2
Perfect timing to catch up on #NeurIPS2023 🤓
Here's how you can easily build an ArXiv research assistant 🧑🔬🤖 with @llama_index + LionAGI by @quantoceanli 👇 1) Define a @llama_index RAG pipeline as an agent tool 2) Plug tool into LionAGI, letting you compose custom agentic pipelines Check it out: https://t.co/GaSf1ie59N
0
0
0
awesome. was thinking of @StepsizeHQ just yesterday. integrated tools to reduce technical debt decision making.
1
1
4
Couldn't Be Better... Stepsize — The only issue tracker for software engineers Thnx @matthl @NickOmeyer @mutlu82 @AlexOmeyer
producthunt.com
Stepsize is an editor-first issue tracker for a healthier codebase. Create, prioritise, and fix code issues directly from your editor — perfect for tech debt and refactoring. Integrates with Jira,...
0
1
2
As a developer, you need to organise your tasks to manage your time and successfully complete a sprint. These 6 VS Code extensions help devs get organised and work effectively. { author: @StepsizeHQ } #DEVCommunity
https://t.co/9pk5ulqcyW
0
6
28