Polly is now embedded across every component of LangSmith—tracing, experiments, and datasets. An assistant icon appears in the lower‑right corner of any window and retains session context, so when you switch tasks it remembers what you were working on.
The AI assistant does more than answer questions; it takes action. It updates prompts, generates datasets from failed runs, writes evaluator code, and filters projects. In practice, Polly functions like an engineering consultant, applying the needed fixes instantly.
The expected return on investment is up to 20% thanks to accelerated debugging, reduced manual corrections, and less development downtime. Automating routine tasks frees the team to deliver real business value instead of digging through endless traces.
For you, this means you can ship features faster, cut costs, and gain a competitive edge in AI products.
Why this matters: Faster iteration shortens time‑to‑market while automation drives cost efficiency. Executives should deploy Polly now to unlock higher ROI and keep their AI offerings ahead of the competition.