Hi Friends! 👋

📊 Daily Stand-Up: Yesterday, as the Lead Engineer, we worked on our integration with AI. We quickly ran into the problem that in order to play our game with the AI, we needed to seed it with information that it can use against the player, information we generate. In a normal chat scenario, you have the assistant/system and the user/human and if we give all the AI that information as the user the game falls apart. We’ve tried various methods to circumvent this and it has worked to various degrees. We need a communication framework that allows for us to swap out Ollama for a hosted service. So what do we do…

Introducing Langchain w/ Ollama. Langchain is a framework ported to several languages that can abstract the communication with whichever AI we want. We’ve so far set it up to wrap our llama2 AI. But back to our problem, we need to seed it with information. In comes RAG (Retrieval Augmented Generation). This is exactly what we need, an AI that has access to some generated information that we can then interact with.

Today we’re diving into setting up our Langchain RAG with Ollama.

💪🏼 New Skill: Langchain & Retrieval Augmented Generation

🎵 Playlist: https://open.spotify.com/track/0DANcJuMamcL9NyYkEWWTq?si=07a9fce035fe453b

🧠 Mental State: Frustrated.

🛠️ Tools: MacBook Pro M2, Ipad Pro, Pixel 6, Intellij, Flutter, The Keyboard that Saved my Hands, The Desk that Saved my Back, The Perfect Stand for Ipads, Desk Mounts that Work and Go on Sale, Ollama

📚 Resources: 

https://pub.dev/packages/langchain_ollama

https://github.com/davidmigloz/langchain_dart

#AI, #TechJourney, #Flutter, #MobileGameDevelopment, #SoftwareDevelopment, #ProgrammingLife, #TechBlog

By: David Giametta

Future David here! I hope you enjoyed reading through this post. AI Murders is now live! Apple, Google