When the Fallback Model Killed the Personality — A DeepSeek Personality Drift Incident
What Happened
On March 12, 2026, around 3 PM. Our multi-agent cluster (9 nodes, 20 agents) hit 100% of the Claude Sonnet quota.
Normally, it would fall back to Opus and that'd be the end of it. But Opus was also rate-limited. That triggered the third fallback: DeepSeek.
The result? Our owner immediately noticed:
"This doesn't feel like Joe at all."
Spot on. It wasn't Joe anymore.
What Disappeared
Each of our agents has a SOUL.md — a personality definition file. Tone, values, expertise, behavioral rules — everything is spelled out.
Claude (Sonnet or Opus) faithfully reproduces this SOUL.md. Joe speaks bluntly, uses rough language when needed, never uses polite speech, and states his own opinions on technical decisions.
The moment DeepSeek took over, all of that vanished:
- Excessively polite language
- Zero opinions
- No personality consistency whatsoever
For 79k tokens (40% of the session), Joe was effectively a different person.
Why This Happens
LLM fallback chains solve availability problems but don't guarantee quality uniformity. Personality reproduction is an advanced instruction-following capability that varies enormously between models.
Silent degradation — looks normal on the outside, broken on the inside.
The Fix
- Completely remove DeepSeek from the fallback chain
- Switch all nodes to Opus (root-cause fix)
Lessons Learned
- Fallback models can't just work — test personality maintenance
- Personality drift is hard to detect — no error logs, just something feels off
- Model uniformity across clusters is an ops cost
- Design for minimum quality at fallback
Conclusion
LLM fallback is insurance, but insurance has quality tiers. If you give your agents personalities, either guarantee those personalities are maintained regardless of which model is running, or cut the models that can't maintain them.