Fetching from the wire…
01
02
03
04
05
06
07
08
Vibe Coding2026-07-20 · source-backed
Quesma's post-mortem on a custom deep-research pipeline whose token-optimization effort cost more than it saved. 155 points, 197 comments. The value is in the specifics of where multi-agent research pipelines leak: redundant fan-out, re-fetching the same sources across agents, and context re-injection between stages. I run a pipeline shaped exactly like this and all three leaks are real. Re-fetching across agents is the worst one because it looks like parallelism.
Each link below shares sources, entities, or timing with this story.
Shared topic / Tension / Downstream implication
Overlapping topics (agent, between, context); pushes against this story (against); traces where this leads (implication).
Overlapping topics (agent, context); pushes against this story (against); traces where this leads (downstream).
Shared topic / Tension
Overlapping topics (agent, cost, token); pushes against this story (against).
Overlapping topics (agent, context, cost); pushes against this story (against).
Overlapping topics (agent, cost, token); pushes against this story (versus).
Overlapping topics (agent, between, cost); pushes against this story (vs).
Overlapping topics (agent, between, context); pushes against this story (but).
Overlapping topics (agent, context, token); pushes against this story (versus).