Fede Nolasco, Strategy, AI, Architecture and Data

State of Prompt Engineering

Posted by Fede Nolasco | Oct 28, 2023 | Generative AI

Discover the groundbreaking impact of prompt engineering on AI and task performance. Dive deep into the State of Prompt Engineering and unravel its marvels. Explore techniques like Chain of Thought, Tree of Thought, and Graph of Thought. Learn about the Automatic Prompt Engineer and Optimization by Prompting. Consider the potential of these advancements to revolutionize problem-solving. Embrace the changes and advancements in AI and prompt engineering for a more innovative future.

Does Context Length Matters?

Posted by Fede Nolasco | Oct 27, 2023 | Generative AI, LLM

The stateof. ai 2023 report explores the impact of context length on language models’ performance and the need to increase it. While large language models offer few-shot capabilities, context length can be limiting. Innovations like FlashAttention and ALiBi address this issue. However, longer context lengths don’t always guarantee better performance, as research suggests. Proprietary models also handle longer context lengths better than open models. #AI #LanguageModels

Building a Single User Generative AI application with Python

Posted by Fede Nolasco | Oct 2, 2023 | AI Coding, ChatGPT, Data Architecture, Generative AI

Discover the power of generative AI and learn about a single user AI orchestrator application. This Python-based app showcases core orchestration functions, leverages accelerators like ChatGPT, and provides hands-on experience with generative AI workflows. Find out how to unlock tailored solutions and maximize productivity with effective prompts. Explore the mechanics of the single user app and its modular architecture, including UI, basic orchestrator, API models, Elasticsearch, Python, and Streamlit. Get a personalized and efficient AI experience with this open-source application available on GitHub.