Graham McIntire 9a10e01a24 fix(llm): switch to deepseek-chat (standard model, no reasoning overhead) ... deepseek-v4-flash is still a reasoning variant and burns tokens on chain-of-thought before producing content. deepseek-chat produces output directly — 82 tokens all went to content in testing, vs 85-102 tokens that produced empty observations on flash.		2026-05-11 10:23:18 -05:00
..
mix/tasks	refactor: move temperature backfill to a release-safe module	2026-05-10 12:50:35 -05:00
snmpkit	refactor: convert 6 Gleam modules to idiomatic Elixir with TDD (#196)	2026-03-28 09:52:07 -05:00
towerops	fix(llm): switch to deepseek-chat (standard model, no reasoning overhead)	2026-05-11 10:23:18 -05:00
towerops_web	refactor(webhooks): enqueue Oban jobs immediately, ACK fast	2026-05-10 17:18:23 -05:00
snmp_lib.ex	bring in snmpkit	2026-01-23 12:52:17 -06:00
snmp_mgr.ex	credo fixes	2026-01-23 14:01:52 -06:00
snmpkit.ex	complete overhaul of snmp engine	2026-01-30 10:41:07 -06:00
towerops.ex	init	2025-12-21 11:10:43 -06:00
towerops_native.ex	dialyzer: replace suppressions with real fixes where possible	2026-04-21 11:01:30 -05:00
towerops_web.ex	feat(coverage): write rasters to NFS-backed dir, serve via dedicated static plug	2026-05-07 07:38:05 -05:00