Ingrid Pan @_ingridpan

she/her. product @Gradient_AI_ gradient.ai San Francisco Joined February 2011

Tweets

76
Followers

137
Following

201
Likes

83

Ingrid Pan @_ingridpan

2 years ago

I’ll only be at a 7am breakfast in Vegas for @MongoDB’s GenAI panel

0 0 4 201 0

RAG 101 for Enterprise 🔎🧑‍💼 This overview by @Gradient_AI_ is a fantastic intro to some core concepts around RAG, why you should care, and how to get optimal performance 💡 Open vs. closed-source LLMs 💡 Prompt engineering vs. fine-tuning vs. RAG 💡 Optimizing RAG: chunking,…

LlamaIndex 🦙 @llama_index

2 years ago

0 33 173 153K 160

Download Image

3 98 489 136K 556

Download Image

HoneyHive @honeyhiveai

2 years ago

After recent OpenAI reliability issues and uncertainty surrounding the company's future, many developers are looking to switch to open-source models, but find themselves unprepared. The solution is simple - set up your data flywheel early. Blog Post: honeyhive.ai/post/openai-vs…

0 5 10 2K 2

Download Image

Ingrid Pan @_ingridpan

2 years ago

Landed in Singapore and looking forward to @sgfintechfest #SFF2023 We’re talking AI use cases in finance tomorrow at Elevandi Insights Day! @Gradient_AI_ fintechfestival.sg/agenda/804-bey…

0 0 2 98 0

Download Image

Tony Dinh 🎯 @tdinh_me

2 years ago

It costs $2-3 million to train a custom GPT-4 model with your own dataset.

160 236 4K 1.5M 625

Download Image

Greg Kamradt @GregKamradt

2 years ago

I tried cloning myself with a LLM + video gen, but the hardest part was matching my tone (what & how I say) Here’s my journey that made it happen (the 7th method actually worked) Turns out fine tuning w/ @gradient_ai_ and the right training set made it happen The process:

5 8 54 8K 33

Download Video

DeepSky @DeepSkyAI

2 years ago

There's some talented creators out there. @GregKamradt uses Gradient's fine-tuning capabilities to clone himself in latest video. 😲 Why Gradient? ✅ No need for dedicated infrastructure ✅ Cost Effective ✅ Integrations w/ partners Like @LangChainAI youtube.com/watch?v=dzPsXa…

0 6 16 7K 7

Ingrid Pan @_ingridpan

2 years ago

Sharing a great opportunity for PhD candidates and researchers to pursue LLM development projects! 👩‍🎓 Love that @Gradient_AI_ is supporting innovation in AI 🚀💡

0 1 2 177 0

Ingrid Pan @_ingridpan

2 years ago

Just wrapped @money2020 💼 Enjoyed meeting folks and discussing how LLMs can ✅ augment investment workflows ✅ streamline identity verification ✅ underwrite loans ✅ prevent fraud and more... Looking forward to the next one!

0 0 2 141 0

Jerry Liu @jerryjliu0

2 years ago

We’re launching cookbooks 🧑‍🍳 (in partnership w/ @Gradient_AI_) to let you easily 1) fine-tune an open-source LLM (e.g. Llama 2), and 2) plug into your @llama_index LLM/RAG pipeline - no PyTorch required! ✅ fine-tune nous-hermes ✅ fine-tune llama2 for better text-to-SQL Our…

6 52 317 47K 266

Download Image

Ingrid Pan @_ingridpan

2 years ago

Thanks for coming out to #LWTSummit @lesbiantech and learning about how LLMs are catalyzing the Fourth Industrial Revolution ⚡️ Love the energy and representation today - so glad to be a part of this community!

0 1 7 2K 0

Download Image

Ingrid Pan @_ingridpan

2 years ago

Great to be here today and looking forward to meeting folks! ⚡️ Come find me after the talk if you want to chat. #LWTSummit

LWT @lesbiantech

2 years ago

Great to be here today and looking forward to meeting folks! ⚡️ Come find me after the talk if you want to chat. #LWTSummit

0 2 3 2K 0

0 0 3 266 0

Ingrid Pan @_ingridpan

2 years ago

Really excited about our latest integration with @llama_index 🦙🦙🦙 This means you can use your @Gradient_AI_ fine-tuned LLMs + embeddings seamlessly in your RAG system.

LlamaIndex 🦙 @llama_index

2 years ago

Really excited about our latest integration with @llama_index 🦙🦙🦙 This means you can use your @Gradient_AI_ fine-tuned LLMs + embeddings seamlessly in your RAG system.

1 28 102 15K 58

Download Image

0 1 2 198 1

Greg Kamradt @GregKamradt

2 years ago

Generating realistic audio/video is nearly a solved problem at this point The hard part of cloning yourself is mimicking *what* you say and *how* you say it I'm starting to get some pretty solid results fine tuning nous hermes with @Gradient_AI_ "Tell me about your first job"