alex boquist

[alex boquist @BoquistAlex]

Shedeur first nfl td @ShedeurSanders

[Scout @scoutshq]

🧠 Smarter RAG starts with better data. Learn the basics of web scraping with our comprehensive guide to getting clean, usable data.

[Scout @scoutshq]

🚀 Exciting news! You can now create a @SlackHQ AI Agent with access to @linear. Just mention the bot and watch it handle tasks for you! 🤖💥

#AIAgents #LLM #AgenticAI #SlackBot #ProductivityBoost

[Scout @scoutshq]

Guardrails in AI: How to Protect Your RAG Applications from Inaccuracies:
https://t.co/9jZuzglKyA

[Scout @scoutshq]

Not sure which LLM to use?! Why not use them all! In this simple newsletter generator we're combining the output of 4 different models, allowing you to pull the best parts of each and combine them into a perfect response. #AI #RAG #llama3 #mixtral #openai

[alex boquist @BoquistAlex]

@postgresml @postgresconf save me an article of swag plz

[Scout @scoutshq]

Reliable JSON outputs are 🔑. You can now set your desired JSON schema on your LLM calls in JSON mode!

#LLMs @FireworksAI_HQ 🙏

[Scout @scoutshq]

Sneak peek at the future of Scout

[Chase Lean @chaseleantj]

Big news for logo and graphic designers!

Midjourney had a handicap: text generation.

The words it generated were mostly meaningless.

The new inpainting feature solves this problem.

This guide will show you how to do it:

[Matt Shumer @mattshumer_]

Pro tip:

If your GPT prompt isn't doing what you want it to do, put it into https://t.co/TyrXif9pBC.

It'll show you how the model 'sees' your prompt, and from there, you can improve it!

[Logan Kilpatrick @OfficialLoganK]

The @OpenAI cookbook is one of the most underrated and underused developer resources available today ⭐️

Here are 7 notebooks you should know about 🧵

[Mark Tenenholtz @marktenenholtz]

Best AI advice I can give is to learn how to fine-tune models.

2 years ago, training a model with >1B parameters was a pipe dream for most.

Now, you can:

• Train a 65B model on a single GPU
• Generate w/ 30B+ param models on the CPU

But best of all… (cont)

[Mark Tenenholtz @marktenenholtz]

You probably don't need a context window longer than 8k.

LLMs cannot handle a lot of irrelevant context in their prompt. Even if you give it the right answer in the prompt, it still might get confused.

Great retrieval >>> shoving more tokens into the prompt

[alex boquist @BoquistAlex]

@agishaun @OpenAI awesome! nice work

[Shaun.AGI @agishaun]

After hours of reading, testing and trying @OpenAI function calling, I present you the Best Function Calling template
- it allows you to decorate any function as openai_function
- it guarantees json with schema validations

(gist link attached)
🧵👇
https://t.co/OY8lUDywYD

[Hyung Won Chung @hwchung27]

I've been experimenting with Test Driven Development with GPT-4.

I first write test cases to formalize the desired behavior, then ask GPT-4 to write a function and suggest additional tests if needed.

I've found this method more efficient than writing the function first and then asking GPT-4 for test cases.

This approach lessens mental strain, allowing me to focus on the broader structure of the code without getting mired in implementation details.

It's too early to say, but I believe that as language models continue to improve, more human effort will be directed towards devising effective test cases.

[BigCode @BigCodeProject]

📣 Introducing ⭐ StarCoder+ & StarChat Beta!

We trained StarCoder on the Falcon model's English web dataset and Instruction-tuned it. Both models rank high in the LLM leaderboard, with strong natural language performance and coding capabilities.

https://t.co/sWlPy5ILtK

[Loubna Ben Allal @LoubnaBenAllal1]

Stay tuned ✨!

[DeepLearning.AI @DeepLearningAI]

Building Systems with the ChatGPT API is live!

In this short course, you’ll learn how to break a complex task down to be carried out via multiple API calls to an LLM.

Join for free: https://t.co/diP5XcRU9w

[Statsig @statsig]

🚨Integration alert🚨

This week, we launched the statsig-langchain package in partnership with @hwchase17 and the rest of the @langchain team.

Devs can use this package to set up event logging and experiment assignment in a Langchain application within minutes.

Link below 👇