Thanks for making this series. I'm waking up every morning and watching them before work so please keep them coming. Good learning material. The LMM section with the coffee stain is the best. More of this please! I'll keep giving a thumbs up. How active is your discord community?
Amazing series. Waiting for next episodes. Would also love to see some additional info on how to easily, securely (and ideally with a low cost or free) to run such code with flask apps integrated with chatgpt in the backed not on a local machine but on the web
Fantastic session. I am only a couple of days "messing" around with LLM. I found LM Studio tonight which seems to be a vehicle to many LLMs. When I finish I go to the previous video which for some reason I was not notified, but I'll go get it. The dta analysis portion was realy cool. I'm working on Power BI and a lot of data analysis visualiztion happens with that. I feel pretty good knowing CSS and making some really nice displays. I have some courses I''m taken so I won;t be able to devote the time needed, but this poertion of know how, helps a lot, because nothing iis unfamiliar. I used to just get it a daze and konk out, I didn't nod once. So, I'm off to see the wizard, oh wait, that's you!! I off to go see the previous video. See you later.
Hey man! I wasnt able to find the codes you used in this video, did you not upload them since they were not really mini projects but quick little examples?
Thank you for the video - without LLMs I am not able to code (I have medical education) - GPT-4 is great for Python - I am struggling only with new libraries (for example Langchain or Autogen), other elements are flawless.
Yeah I hope they retrain using more recent data. A lot has happened since 2021, especially in the domain of AI integration programming. And web programming also changes at a fast pace.
1:03: 💻 In this video, the speaker discusses the importance of AI engineers learning to use AI tools for code generation, such as GitHub co-pilot and replateAI, and highlights the benefits of using these tools. 4:39: ✨ The video discusses the benefits of using AI tools for coding, including generating code, providing context, code commenting, debugging help, and translating code between languages. 10:20: 💻 The video discusses the use of coding assistant tools like GitHub Copilot and Replit AI, highlighting their benefits and features. 14:54: 👨💻 The video discusses using Replit AI to create a simple UI for a chatbot and generating code from an LMM model. 19:38: 👨💻 The video demonstrates how to use OpenAI's Codex to generate code from visual inputs and discusses its potential use cases. 24:42: 💻 The video showcases the use of GPT-3 for various Python coding tasks, including generating code, reverse engineering UI, and data analysis. 29:49: 📊 The video showcases the use of OpenAI's GPT-3 in data analysis and visualization, demonstrating its ability to generate Python code and perform tasks like analyzing sales trends and creating compound interest calculators. 40:30: 👨💻 The video discusses using OpenAI's API to generate code and demonstrates how to improve code using dynamic programming. 45:19: 👩💻 The video discusses the use of AI models to generate code and highlights the benefits of open source software. 50:42: 📚 The speaker discusses the importance of practice and experimentation in the field of AI engineering and predicts exciting advancements in the future. Recap by Tammy AI
46:12 He is loading 2-bit quantized (Q2_K) Code Llama 7B into the 16GB GPU he has just bragged about. That's completely misleading. That GPU can fit the 13B model at full 8 bits or at least 6 bits if the OS is using up a bit of VRAM. At 2 bits it can even fit the 34B model, however I would not suggest to do that. Depending on your use case don't go below 5 or 6 bits if you don't want really bad results.
hi! whole Marketing Team working on totally tailored to your own use case Strategy. AutoGEN + MemGPT + Local LLM. can you imagine how powerful this thing might be? 😍 maybe you that advanced with AI Agents, that you might try to do that? would be awesome! kind of like your AI Swarm Worker Agents. maybe next episode? 😀
Amazing series, thanks! It would be great if you can create some with local LLMs and RAG systems (btw, ruclips.net/video/zwGrae75yAg/видео.html, you were right the first time: LLM not LMM :D )
Thanks for making this series. I'm waking up every morning and watching them before work so please keep them coming. Good learning material. The LMM section with the coffee stain is the best. More of this please! I'll keep giving a thumbs up. How active is your discord community?
I love this series! Please proceed, can not wait for the next episode :)
Amazing series. Waiting for next episodes. Would also love to see some additional info on how to easily, securely (and ideally with a low cost or free) to run such code with flask apps integrated with chatgpt in the backed not on a local machine but on the web
Great series!
This video was a great starting point for anyone starting to explore AI engineering skills.
Best Channel about AI!
Fantastic session. I am only a couple of days "messing" around with LLM. I found LM Studio tonight which seems to be a vehicle to many LLMs. When I finish I go to the previous video which for some reason I was not notified, but I'll go get it. The dta analysis portion was realy cool. I'm working on Power BI and a lot of data analysis visualiztion happens with that. I feel pretty good knowing CSS and making some really nice displays.
I have some courses I''m taken so I won;t be able to devote the time needed, but this poertion of know how, helps a lot, because nothing iis unfamiliar. I used to just get it a daze and konk out, I didn't nod once. So, I'm off to see the wizard, oh wait, that's you!! I off to go see the previous video. See you later.
Hey man! I wasnt able to find the codes you used in this video, did you not upload them since they were not really mini projects but quick little examples?
Thanks for the video! Learning a lot.
Great video!! Keep it up
Thank you for the video - without LLMs I am not able to code (I have medical education) - GPT-4 is great for Python - I am struggling only with new libraries (for example Langchain or Autogen), other elements are flawless.
I just started learning python last month... ChatGPT definitely help a lot.
Yeah I hope they retrain using more recent data.
A lot has happened since 2021, especially in the domain of AI integration programming. And web programming also changes at a fast pace.
Great video, thanks for your efforts. learn a lot!
Good video, thanks
Can someone give use cases for someone who mostly uses chatgpt for creation of better documents and brainstorming?
Amazing video! Can you also share the persona prompt used in Custom Insttuctions?
1:03: 💻 In this video, the speaker discusses the importance of AI engineers learning to use AI tools for code generation, such as GitHub co-pilot and replateAI, and highlights the benefits of using these tools.
4:39: ✨ The video discusses the benefits of using AI tools for coding, including generating code, providing context, code commenting, debugging help, and translating code between languages.
10:20: 💻 The video discusses the use of coding assistant tools like GitHub Copilot and Replit AI, highlighting their benefits and features.
14:54: 👨💻 The video discusses using Replit AI to create a simple UI for a chatbot and generating code from an LMM model.
19:38: 👨💻 The video demonstrates how to use OpenAI's Codex to generate code from visual inputs and discusses its potential use cases.
24:42: 💻 The video showcases the use of GPT-3 for various Python coding tasks, including generating code, reverse engineering UI, and data analysis.
29:49: 📊 The video showcases the use of OpenAI's GPT-3 in data analysis and visualization, demonstrating its ability to generate Python code and perform tasks like analyzing sales trends and creating compound interest calculators.
40:30: 👨💻 The video discusses using OpenAI's API to generate code and demonstrates how to improve code using dynamic programming.
45:19: 👩💻 The video discusses the use of AI models to generate code and highlights the benefits of open source software.
50:42: 📚 The speaker discusses the importance of practice and experimentation in the field of AI engineering and predicts exciting advancements in the future.
Recap by Tammy AI
46:12 He is loading 2-bit quantized (Q2_K) Code Llama 7B into the 16GB GPU he has just bragged about. That's completely misleading. That GPU can fit the 13B model at full 8 bits or at least 6 bits if the OS is using up a bit of VRAM. At 2 bits it can even fit the 34B model, however I would not suggest to do that. Depending on your use case don't go below 5 or 6 bits if you don't want really bad results.
hi!
whole Marketing Team working on totally tailored to your own use case Strategy. AutoGEN + MemGPT + Local LLM. can you imagine how powerful this thing might be? 😍
maybe you that advanced with AI Agents, that you might try to do that?
would be awesome!
kind of like your AI Swarm Worker Agents. maybe next episode? 😀
Microservices architecture and using grpc to communicate and kubernetes orchestration with code running in docker containers
Thansk 😮
Amazing series, thanks! It would be great if you can create some with local LLMs and RAG systems (btw, ruclips.net/video/zwGrae75yAg/видео.html, you were right the first time: LLM not LMM :D )