Mark your calendars for November 15th and bring a friend—it’s time for another exciting Pyerevan meetup! Let’s explore cutting-edge Python topics, share ideas, and connect with our amazing community.

This time, we’re thrilled to host two incredible speakers:

Aleksei will talk about "Building LLM-Based Services with Your Own Inference", where he’ll share insights on deploying large language models (LLMs) on your own GPU infrastructure without relying on external APIs.

• Learn how to optimize costs, enhance data privacy, and perform inference with pre-trained or fine-tuned models.
• Topics include core components for deploying LLM services, selecting the right model, and creating APIs for LLMs with the right inference engine.

Erik will present "How to Reason Robustly, Verifiably, and Faithfully with Large Language Models", diving deep into methods for enhancing reasoning capabilities in LLMs.

• Explore challenges like faithfulness, verifiability, and interpretability in LLMs.
• Learn about Chain-of-Thought (CoT) prompting, context-aware decoding, and symbolic reasoning approaches such as FLARE, to advance task-faithful and interpretable AI reasoning systems.

No change in the rules: everyone is welcome, regardless of your tech stack or experience level. The event is free, but registration is required.