大模型、多模态、推理模型和 AI 产品趋势
This collection features articles and community discussions in the AI 前沿 space, focusing on real-world engineering scenarios, tool selection, practical paths, and risk boundaries.
The page is continuously updated and serves as a learning roadmap, technology selection reference, and technical review entry point.
Putting documents in a vector database and using top-k retrieval is only RAG's "hello world." This article covers four levers for improving RAG quality: chunking, hybrid retrieval, reranking, and query rewriting, plus a reproducible evaluation method that takes RAG from "barely usable" to genuinely accurate.
Covers tech trends, engineering practices, and indie developer perspectives around large models, RAG, and AI.
Technical trends, engineering practices, and an independent developer's perspective on LLMs and open source.
Technical trends, engineering practices, and an independent developer's perspective on RAG and large language models.
Around the technical trends, engineering practices, and indie-developer perspectives of large models and AI.
Technical trends, engineering practices, and an independent developer perspective on RAG and large language models.
Covers tech trends, engineering practices, and indie developer perspectives around large models and RAG.
A high-quality prompt is not a magic instruction but an engineering system comprising a task contract, context, examples, an output structure, and an evaluation set. This article presents a reusable prompt design method and explains how version control and regression testing enable stable iteration.
"Model leaderboards cannot replace business evaluation. This article designs a five-layer LLM application evaluation framework spanning evaluation-set construction, rule-based metrics, model judges, human review, and production observation to support model selection, Prompt regression, and RAG optimization."
Not every LLM task needs to run in real time. Batch classification, summarization, labeling, and content generation can cost half as much with Batch APIs. This article covers OpenAI and Anthropic Batch APIs, self-hosted batching with vLLM, job pipelines, and real-time vs. batch tradeoffs.
For RAG and semantic search, choosing a vector database is unavoidable. This article compares pgvector, Qdrant, Milvus, Chroma, and Pinecone by data scale, latency, operational cost, and hybrid search, with clear recommendations for when to use each.
Saving the model name, Prompt, and final score is not enough to replay an evaluation. This article freezes data and label generations, the request sequence, the decoder, the runtime environment, the Evaluator, and the result set to form a verifiable run evidence root.