​Join us for a hands-on session diving into advanced techniques for building and optimizing AI / LLM applications.

​​Co-hosted by: Boundary X AI Tinkerers X AWS Gen AI Lofts

​​What to Expect 🚀

​​This interactive session will include live coding, technical explanations, and Q&A. 

​​We’ll explore various topics including -

​​Maximizing LLM performance: 

• ​​Improve Accuracy: Techniques to boost the reliability of model outputs and reduce hallucinations. How to go from 90% to 99%? 

• ​​Improve Consistency: How can your customers get the same response every single time?

• ​​Manage Latency: Practical methods for reducing delays in your AI pipeline.

• ​​Eval techniques: What works beyond LLMs as a judge? Can LLMs even judge their own outputs?

​​Challenges devs face when working with LLM:

• ​​Why are LLMs still non-deterministic even when temperature == 0 or a seed is set?

• ​​How to manage and scale your codebase when you have hundreds of prompts that need to be managed by engineers and non-engineers?

• ​​Why are log probabilities unreliable as a confidence metric?

• ​​What is wrong with token-based streaming and how semantic streaming is better.

​​Key Details 🗓️

• ​​Date: December 19, 2024 (Thursday)

• ​​Time: 2:00 PM - 5:00 PM 

• ​​Location: AWS 19th floor - 525 Market St, San Francisco, CA 94105

​​Agenda ⏰

• ​​2:00 PM: Presentation w/ live coding

• ​​3:30 PM: Q&A

• ​​4:30 PM: Wrap-up networking

​​Who is this for? 🤖

​​This event is specifically curated for:

• ​​Developers, AI engineers, and data scientists that are actively working on AI pipelines.

• ​​Ideally, you should have already used an LLM API programmatically in your codebase, or have built server-client architectures before.

• ​​Prior experience with getting structured output from AI pipeline is nice-to-have.

​​Bring your laptop, as you’ll have the opportunity to code along and apply these techniques in real-time!

​​Meet the Speaker 🧑‍💻

​​Meet Vaibhav Gupta, one of the creators of BAML. He has worked almost 10 years in AI performance optimization at places like Google, Microsoft, and D. E. Shaw. He loves diving deep and chatting about anything related to Gen AI and Computer Vision! 

​​What is BAML and Boundary?

​​BAML is a domain-specific language to generate structured outputs from LLMs.

​​The team at Boundary created BAML to help developers build reliable Agents, Chatbots with RAG, data extraction from PDFs, and other AI apps.

​​What is AI Tinkerers?

​​AI Tinkerers is a global network of passionate AI practitioners working to push the boundaries of language models and generative AI.

​​What is AWS Gen AI Lofts?

​AWS GenAI Lofts are collaborative spaces and immersive experiences for startups and developers.