Nexus Tech Lead @NexusTechLead

Driving innovation in AI and tech leadership | Passionate about shaping the future of software development with AI insights. Innovation Nexus Joined January 2024

Tweets

35
Followers

40
Following

118
Likes

10

Sumanth @Sumanth_077

9 months ago

AWS released a new Multi-Agent AI framework! A flexible and powerful framework for managing multiple AI agents and handling complex conversations. It let's you dynamically route LLM queries to the most suitable agent based on context. 100% Open Source.

28 325 2K 181K 3K

Download Image

Addy Osmani @addyosmani

9 months ago

Quickly turn a GitHub repository into text for LLMs with Gitingest ⚡️ Replace "hub" with "ingest" in any GitHub URL for a text version of the codebase.

63 395 3K 283K 4K

Download Video

Greg Kamradt @GregKamradt

a year ago

Details on @OpenAI's new assistants RAG *Hard* creep into vectorstore territory Thoughts: * Default chunk overlap of 50%, super interesting * Metadata filtering, super interesting how this dips into vectorstore territory * Unsure about what chunking method they use - 800 tokens…

40 120 904 245K 1K

Download Image

Andrew Ng @AndrewYNg

a year ago

Tool use, in which an LLM is given functions it can request to call for gathering information, taking action, or manipulating data, is a key design pattern of AI agentic workflows. You may be familiar with LLM-based systems that can perform a web search or execute code. Some of…

80 302 2K 256K 1K

carlos @_carlosejimenez

a year ago

SWE-Agent is an open-source software engineering agent with a 12.3% resolve rate on SWE-Bench! Check out SWE-agent in action at swe-agent.com Repo: github.com/princeton-nlp/…

30 110 553 138K 471

Download Video

Matthew Berman @MatthewBerman

a year ago

Devin clones are coming quickly. I finally got Devika working and was able to reproduce the "game of life" demo easily. Here's a full tutorial:

13 52 358 27K 243

Download Video

Matt Shumer @mattshumer_

a year ago

Here is an insanely useful Claude 3 prompt for engineers. Use it to automatically generate unit tests for your code: --- <prompt_explanation> You are an expert software tester tasked with thoroughly testing a given piece of code. Your goal is to generate a comprehensive set of…

38 202 2K 276K 4K

Download Image

Andrew Ng @AndrewYNg

a year ago

Last week, I described four design patterns for AI agentic workflows that I believe will drive significant progress this year: Reflection, Tool use, Planning and Multi-agent collaboration. Instead of having an LLM generate its final output directly, an agentic workflow prompts…

101 576 3K 488K 3K

Jerry Liu @jerryjliu0

a year ago

A nice analogy of RAG vs. finetuning is to compare it to an open vs. closed-book exam 📖RAG == open-book exam without studying. Only use information provided in the page, but hard to discern which information is relevant. 📘Finetuning == closed-book exam. Only use memorized…

LlamaIndex 🦙 @llama_index

a year ago

15 109 518 113K 520

Download Image

13 113 587 68K 435

Download Image

Matt Shumer @mattshumer_

a year ago

Useful Claude 3 trick to help you visualize code better. Paste some code in, and ask it to make a flowchart. Then, paste the flowchart code into a Mermaid viewer, and you'll get a nice, understandable visualization of your code!

37 138 1K 122K 1K

Download Image

João Moura @joaomdmoura

a year ago

What can you do with AI Agents? Let me give you a thought framework to answer that question :😎👉👉

6 9 116 7K 45

Download Video

Nexus Tech Lead @NexusTechLead

a year ago

AI Agents for Platform Upgrades - makes 100% sense

Andrew Ng @AndrewYNg

a year ago

AI Agents for Platform Upgrades - makes 100% sense

217 1K 5K 831K 4K

Download Image

0 0 1 31 0

Jason Arbon 🧙‍♂️ @jarbon

a year ago

The looming 100X testing problem. GenAI will soon create 10X more code, 10X faster. Human testers will have to partner with AI to try and keep up.

1 2 5 321 0

Martin Fowler @martinfowler

2 years ago

NEW POST I'm hearing of more teams working in a remote-first style. But many are making a mistake by never meeting face-to-face. Effective remote-first teams I've encountered do periodic gatherings 2-4 times a year. martinfowler.com/bliki/Periodic…

10 71 440 54K 137

Jerry Liu @jerryjliu0

a year ago

If you want to build advanced RAG/agents over your codebase, the first step is a good code parsing algorithm. This is an awesome 🧵 by @akshay_pachaar that shows you how to use the @llama_index CodeHierarchyNodeParser to build a knowledge graph over your codebase - it…

Akshay 🚀 @akshay_pachaar

a year ago

13 61 305 82K 279

Download Video

7 64 353 45K 350

Download Video

Ruben Hassid @RubenHssd

a year ago

I don't use ToT prompting enough. But I heard chatgpt is much better with it. So I ran a little experiment: → simple prompt vs. Tree of Thoughts Read my entire benchmark below:

10 31 208 49K 427

Download Image

LangChain @LangChainAI

a year ago

Is RAG Really Dead? Testing the retrieval limits of long context LLMs RAG systems often retrieve multiple docs relevant to a user input and reason over them to return an answer. How well can long context LLMs do this across hundreds or thousands of pages of input tokens?…

10 63 399 56K 331

Download Image

Deedy @deedydas

a year ago

Devin will change the game, but remember: "Per task, 10-20 maxed out GPT4-32k calls /min, thats ~$2-$5/min, or $120-$300/hr depending on Input/Output token ratio." "took about 19mins for it to ask me for the API key from when I started." They crushed orchestration tho.