Skip to content
jongkwan.dev

Posts

31 posts
AllDevelopment#agent#backend#security#infra#data#architecture#llm#insight#rag#algorithm
Reset filters

  1. The Agent Memory Trilemma

  2. Judging Multi-Agent Adoption on Reliability

  3. RAG Moves From Retriever to Compiler

  4. Reranker Objective Shift

  5. Speculative Tool Execution for Agents

  6. The Tool Layer as the Agent Bottleneck

  7. MCP Tool Discovery and Dynamic Loading

  8. Context, State, Memory, and Checkpoints

  9. The Confused Deputy and Agent Identity

  10. Late Interaction Reranking

  11. LLM Agent Memory Poisoning

  12. Evaluating Multi-Agent Systems and Choosing an Orchestration Topology

  13. Agent Harnesses and Loop Engineering

  14. MCP Tool Poisoning

  15. OpenAI o1 and DeepSeek-R1 Reasoning Models

  16. RAG Cross-Encoder Reranking

  17. DPR-Based Dense Retrieval for RAG

  18. Defenses at the Agent, RAG, and MCP Layers

  19. Agent Orchestration with LangGraph

  20. Test-Time Compute Scaling

  21. Toolformer

  22. RAG

  23. MADKE (Multi-Agent Debate with Knowledge Enhancement)

  24. Multi-Agent Debate

  25. LATS (Language Agent Tree Search)

  26. Tree of Thought (ToT)

  27. CRITIC

  28. Self-Refine

  29. Reflexion

  30. ReAct (Reasoning + Acting)

  31. Chain of Thought (CoT) Step-by-Step Reasoning