Blogs
Our technical challenges, ideas, how we're building them, and what we're learning along the way.
π A better Python cache for slow function callsβπ§ Our simple vector database implementationβπ Releasing our Open Source LLM Framework + Playground!βπ Large Language Models are Bad at Refactoring CodeβπΈοΈ A Simple Proxy for Azure and OpenAI raised our GPT4 TPM limit by 24xβπ§ͺ Having GPT-4 Iterate on Unit Tests like a Humanβπ Zero Downtime Deployment with < 50 lines of bashβπΊοΈ How an AI plans code changes across an entire GitHub Repositoryβπ οΈ AI Junior Dev meets Super-Linterβπ€ Why getting GPT-4 to modify files is hardβπΈ When you should handle tech debtβπ Self Hosting an AI Junior Developerβπ Improving LlamaIndexβs Code Chunker by Cleaning Tree-Sitter CSTsβπ How our AI junior dev reads all of your documentationβπ§Ή Sweep's Core AlgorithmββοΈ Generating 50k+ embeddings in 25 seconds using GTE π§ and MapReduce π»βπͺ Chunking 2M+ files a day for Code Search using Syntax Treesβπ€ Letting an AI Junior Dev run GitHub Actionsβπ Code Search Infra for an AI junior developer - that doesn't store codeβπ Understanding your codebase with ctagsβποΈ Building a code search engine in one dayβπ GPT 32k, π» Open-source, DeepLake ποΈ, GPT Functions Search π and more!β