A town-clerk intern copies the same headers, retry loop, and message assembly into every feature branch for chatbot, summarizer, and extractor work. What should a reviewer ask for instead?
Select an answer to reveal the explanation.
Short Explanation
The same headers, retry loop, and message assembly should live in one shared LLM client. Copying that HTTP block into every feature branch guarantees drift. A raw socket per feature, or a serving stack per feature, does not fix the shared-call problem.
Full Explanation
Chatbot, summarizer, and extractor features should share one LLM client or chain so retries, headers, and templates live in one place. Copying the same HTTP block into every branch guarantees drift under review. Separate sockets or a serving stack per feature add cost without fixing the shared-call problem.