My Courses
Career Essentials in Generative AI by ADaSci
Master Generative AI with ADaSci’s comprehensive suite of courses, designed to elevate your career in the rapidly evolving field of artificial intelligence.
How to Integrate LangChain and Hugging Face for Open-Source LLM Deployment?
Discover how the langchain-huggingface package enhances NLP by integrating Hugging Face models with LangChain’s framework.
Analysing Generative AI Embeddings with Azure AI Hub: A Hands-on Guide
Explore and implement various embedding models in Azure AI Hub to enhance contextual search and query responses.
A Hands-on Guide to Enhance RAG with Re-Ranking
Learn how re-ranking in Retrieval-Augmented Generation boosts relevance, enhancing summarization and question answering accuracy.
Building Knowledge Graphs to Enhance Data-Driven Decision-Making
Knowledge graphs, built using graph databases, capture data relationships for efficient modelling and reasoning. This article explores their importance in enhancing data-driven decision-making.
LangChain Vs LlamaIndex for Advanced Query Retrieval
LangChain’s “MultiQuery Retriever” and LlamaIndex’s “Multi-Step Query Engine” enhance advanced query retrieval by ensuring precise, context-aware responses.
Creating Neural Network Architecture Drawings with NN-SVG
NN-SVG is a powerful tool for creating parametric Neural Network architecture drawings, allowing easy export to SVG files for academic papers and web use.
Convert Images of Equations into LaTeX Code Using Python
Convert images of equations into LaTeX code effortlessly with the pix2tex Python library, streamlining the documentation process for researchers and students.
OpenAI’s New Guide on Prompt Engineering: Six Strategies for Better Results
OpenAI’s new Prompt Engineering guide outlines six strategies to improve large language model results, from writing clear instructions to testing changes systematically.
Speeding Up LLM Inference with Microsoft’s LLMLingua
Microsoft’s LLMLingua reduces LLM inference costs and boosts performance by up to 20x with minimal performance loss using prompt compression.
