📄 文章列表
377 篇文章
TDS
Prefill Is Compute-Bound. Decode Is Memory-Bound. Why Your GPU Shouldn’t Do Both.
TDS
Your Chunks Failed Your RAG in Production
TDS
6 Things I Learned Building LLMs From Scratch That No Tutorial Teaches You
TDS
Proxy-Pointer RAG: Structure Meets Scale at 100% Accuracy with Smarter Retrieval
TDS
I Replaced GPT-4 with a Local SLM and My CI/CD Pipeline Stopped Failing
TDS
Using a Local LLM as a Zero-Shot Classifier
TDS
I Built an AI Pipeline for Kindle Highlights
TDS
Bytes Speak All Languages: Cross-Script Name Retrieval via Contrastive Learning
TDS
How to Build an Efficient Knowledge Base for AI Models
TDS
The Must-Know Topics for an LLM Engineer
TDS
Proxy-Pointer RAG — Structure-Aware Document Comparison at Enterprise Scale
TDS
Hybrid Search and Re-Ranking in Production RAG
TDS
Building an Evaluation Harness for Production AI Agents: A 12-Metric Framework From 100+ Deployments
TDS
I Built the Same B2B Document Extractor Twice: Rules vs. LLM
TDS
LLM Evals Are Based on Vibes — I Built the Missing Layer That Decides What Ships
TDS
Can LLMs Replace Survey Respondents?
TDS
EmoNet: Speaker-Aware Transformers for Emotion Recognition — and What I’d Build Differently in 2026
TDS
Five Questions About Chronos-2, the Time Series Foundation Model
TDS
Baseline Enterprise RAG, From PDF to Highlighted Answer
TDS