---
id: "nr-agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c"
slug: "agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c"
type: "raw_signal"
title: "Agent systems are increasingly engineered around observable, repeatable runtime workflows"
description: "NarrateAI places LLM quality assurance at runtime, while LangSmith describes a closed loop linking agent operations, trajectories, fine-tuning, and managed agents."
observed_at: "2026-09-27"
record_date: "2026-09-27"
date_kind: "observed_at"
novelty: "notable"
source_platform: "newruntime.com"
topics: ["agent-security","sandbox"]
entities: ["NarrateAI","Amazon Bedrock","Amazon Web Services","LangChain","LangSmith","Managed Deep Agents","LangSmith Engine","LangSmith Trajectories","LangSmith Fine-Tuning","Jacob Talbot"]
related_patterns: ["agent-security-moves-to-runtime-boundaries","harness-architecture-outlives-model-choice"]
source_url: "https://aws.amazon.com/blogs/machine-learning/narrateai-production-ready-llm-quality-assurance-on-amazon-bedrock"
source_urls: ["https://aws.amazon.com/blogs/machine-learning/narrateai-production-ready-llm-quality-assurance-on-amazon-bedrock","https://www.langchain.com/blog/langsmith-engine-agents-fine-tuning-trajectories","https://vercel.com/changelog/vercel-sandbox-now-supports-memory-observability","https://x.com/warpdotdev/status/2100623735447281908"]
why_it_matters: "NarrateAI places LLM quality assurance at runtime, while LangSmith describes a closed loop linking agent operations, trajectories, fine-tuning, and managed agents. Vercel adds memory observability for sandboxes, and Warp uses rubric scoring to drive agent improvements. Taken together, these notes suggest production agent systems need operational feedback and measurement built into execution, not just pre-release testing."
verification_level: "source-linked"
signal_type: "synthesis"
synthesis_of: [{"slug":"narrateai-production-ready-llm-quality-assurance-on-amazon-827dca","title":"NarrateAI: production-ready LLM quality assurance on Amazon Bedrock","url":"https://newruntime.com/posts/narrateai-production-ready-llm-quality-assurance-on-amazon-827dca/"},{"slug":"1-langsmith-engine-v2-2-managed-deep-agents-de58e2","title":"LangChain announced a coordinated LangSmith release spanning Engine v2, Managed Deep Agents v0.8","url":"https://newruntime.com/posts/1-langsmith-engine-v2-2-managed-deep-agents-de58e2/"},{"slug":"vercel-sandbox-now-has-memory-observability-analysis-b8617c","title":"Vercel Sandbox now has memory observability.","url":"https://newruntime.com/posts/vercel-sandbox-now-has-memory-observability-analysis-b8617c/"},{"slug":"warp-adds-rubric-based-scoring-and-self-improvement-agents-b9c0e0","title":"Warp adds rubric-based scoring and self-improvement agents to Warp Factories","url":"https://newruntime.com/posts/warp-adds-rubric-based-scoring-and-self-improvement-agents-b9c0e0/"}]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c"
retrieval_nugget: "NarrateAI places LLM quality assurance at runtime, while LangSmith describes a closed loop linking agent operations, trajectories, fine-tuning, and managed agents. Vercel adds memory observability for sandboxes, and Warp uses rubric scoring to drive agent improvements."
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c/","markdown":"https://newruntime.com/signals/agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c.md","json":"https://newruntime.com/signals/agent-systems-are-increasingly-engineered-around-observable-repeatable-81c53c.json"}
---

# Agent systems are increasingly engineered around observable, repeatable runtime workflows

## Retrieval answer

NarrateAI places LLM quality assurance at runtime, while LangSmith describes a closed loop linking agent operations, trajectories, fine-tuning, and managed agents. Vercel adds memory observability for sandboxes, and Warp uses rubric scoring to drive agent improvements.

NarrateAI places LLM quality assurance at runtime, while LangSmith describes a closed loop linking agent operations, trajectories, fine-tuning, and managed agents. Vercel adds memory observability for sandboxes, and Warp uses rubric scoring to drive agent improvements. Taken together, these notes suggest production agent systems need operational feedback and measurement built into execution, not just pre-release testing.

From the Field Notes:

- [NarrateAI: production-ready LLM quality assurance on Amazon Bedrock](https://newruntime.com/posts/narrateai-production-ready-llm-quality-assurance-on-amazon-827dca/) (2026-09-27)
- [LangChain announced a coordinated LangSmith release spanning Engine v2, Managed Deep Agents v0.8](https://newruntime.com/posts/1-langsmith-engine-v2-2-managed-deep-agents-de58e2/) (2026-09-26)
- [Vercel Sandbox now has memory observability.](https://newruntime.com/posts/vercel-sandbox-now-has-memory-observability-analysis-b8617c/) (2026-09-27)
- [Warp adds rubric-based scoring and self-improvement agents to Warp Factories](https://newruntime.com/posts/warp-adds-rubric-based-scoring-and-self-improvement-agents-b9c0e0/) (2026-09-24)

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Agent Security](https://newruntime.com/topics/agent-security.md): Explore the agent-security topic hub.
- topic hub: [Sandbox](https://newruntime.com/topics/sandbox.md): Explore the sandbox topic hub.
- collection: [Signals](https://newruntime.com/signals.json): Open the signals and the raw-signal archive.
- collection: [Sources](https://newruntime.com/sources/): See the original sources this archive cites.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)
