TS Compiler Graph 도입으로 LLM 토큰 소모량 10배 절감
I Made TS Compiler Graph MCP: 10x Fewer Tokens in Claude Code
I Made TS Compiler Graph MCP: 10x Fewer Tokens in Claude Code
DeepSeek's Response API Isn't OpenAI Responses. That One Parser Mistake Drops the Reasoning.
5 prompt engineering techniques to get the best out of a legacy project
Prompt injection is role confusion, and your MCP gateway can't see it
Claude Code의 추론 은닉 구조 분석: Signature 기반의 Reasoning 증류 방지 설계
My 7-step prompt chain failed silently at step 6. Here's what replaced it.
VibeThinker: A 3B-Parameter Model Just Beat Opus 4.5 on Reasoning — Here is How
Mastering the Art of LLM Prompting: A Developer's Guide to Getting Better Answers from AI
You Know Zero-Shot, One-Shot & CoT Prompting. But Do You Know ReAct?
Gemini 3.5 Pro: 2M Context, Deep Think, and the Post-Fable-5 Frontier
Presentation: AI Agents to Make Sense of Data at OpenAI
Few-Shot Learning with LLM: A Deep Dive
JSON or XML Tags for LLM Output: The Format That Holds Under Pressure
Can LLMs save themselves from verbosity?
7 Magic Words That Make Your LLM 10 Smarter at Math
Part 6 of 6: How to Build Pipelines That Don't Gaslight Themselves.
SRE 자율성 L0~L4 설계 및 Safety Trifecta를 통한 MTTM 44% 단축
Live chain-of-thought in a chatbot: how to actually stream the tool calls (not just the text)
Types of Prompting: Complete Guide to Prompting Techniques
I built a tool that catches misleading charts using Gemma 4 running locally