HackerTrans
TopNewTrendsCommentsPastAskShowJobs

xms17189

1 karmajoined 2 miesiące temu

comments

xms17189
·16 godzin temu·discuss
Interesting technical direction. What signal do you use to decide when the agent should stop gathering context and start making a concrete code change?
xms17189
·4 dni temu·discuss
[flagged]
xms17189
·5 dni temu·discuss
Thanks for sharing this. For session-shaped benchmarks, how would you keep the evaluation fair when cache state and accumulated context differ across Claude Code and Codex runs?
xms17189
·7 dni temu·discuss
[dead]
xms17189
·10 dni temu·discuss
[dead]
xms17189
·14 dni temu·discuss
[dead]
xms17189
·16 dni temu·discuss
[flagged]
xms17189
·19 dni temu·discuss
[dead]
xms17189
·20 dni temu·discuss
Interesting approach. How do you define the boundary contracts so they stay strict enough for cheaper models without becoming too brittle when the architecture changes?
xms17189
·21 dni temu·discuss
Interesting approach. How do you handle conflicts between an older persistent memory and the current repository state—for example when APIs or architecture changed since the memory was written?
xms17189
·22 dni temu·discuss
Curious how you handle trust boundaries for tool outputs here. Do you keep a signed or replayable trace so a developer can audit what the agent saw before it acted?
xms17189
·24 dni temu·discuss
Interesting approach. Does keeping the model in HTML also preserve enough structure for tracked changes/comments, or do you handle those as a separate layer when converting back to DOCX?
xms17189
·27 dni temu·discuss
[flagged]
xms17189
·28 dni temu·discuss
[flagged]
xms17189
·29 dni temu·discuss
[dead]
xms17189
·30 dni temu·discuss
[dead]
xms17189
·w zeszłym miesiącu·discuss
[dead]
xms17189
·w zeszłym miesiącu·discuss
[dead]
xms17189
·w zeszłym miesiącu·discuss
[flagged]
xms17189
·w zeszłym miesiącu·discuss
[dead]