The authors argue that aligning agent scope with human review capabilities is the key to extracting tacit knowledge and building reliable long-horizon agents.
The round, led by CapitalG, supports LangSmith, which over 7,000 organizations use to run and govern agents.
The agent generated 1,540 artifacts in 12 weeks, saving 337 hours of rep time by using business context rather than just prompt tuning.
A team of four engineers built the platform on Workday's infrastructure, which has since grown to roughly 30 members.
The method uses a three-step process of defining pass/fail criteria, building datasets from real runs, and comparing baseline scores to detect regressions in agent behavior.
The feature is available in Python and JavaScript, with task notifications currently in beta.
The company outlines three principles for using human-in-the-loop workflows to define quality, evaluate context, and create golden datasets for continuous agent improvement.
LangChain manages the data plane infrastructure while keeping traces, evals, and agent deployments in the customer's own cloud account.
By adopting Stagehand's code mode and snapshot features, LangChain reduced per-task browser costs from $26.71 to $1.98 and improved success rates to 90% on WebVoyager.
The shift to background, goal-oriented agents introduces unsolved problems in proactivity, memory, and alignment that standard coding agent patterns do not address.
The new browser tools, built on Stagehand, outperformed six other toolsets in benchmarks, with the closest competitor costing 9x more.
Adding an evidence graph to the OpenWiki memory layer improved accuracy by 27% and cut cost per correct answer by 58%.
The Schedules SDK allows developers to define static or dynamic cron jobs, enabling agents to perform background tasks like memory consolidation and smart triage.
The design agent uses a sandboxed workspace and async sub-agents to generate code-based assets, with 75% of requests coming from outside the marketing department.
The framework requires agents to live in Slack, handle identity via on-behalf-of or fixed credentials, use scoped memory, execute code in sandboxes, and run scheduled background tasks.
The post outlines strategies to prevent context window bloat and stale data by using targeted search tools, offloading large results, and delegating reading tasks to subagents.
Numan uses LangGraph and LangSmith to test its AI health coach against regulatory safety lines, reducing frontier model migration time from weeks to a single day.
The October 2026 update introduces semantic trace search, red-teaming capabilities, and FastMCP support in LangChain v1.4.0.
The LangSmith team details strategies for agent evaluation, including using cheaper judges like Jev to cut costs from $28.17 to $0.34 per run.
The platform is deployed across 120,000 Orange employees and adopted by over 150 customers in France and Europe.
AI Search is available in public beta on LangSmith Cloud at no additional cost, enabling semantic queries across agent traces.
ScaleDown models offer up to 467x cost savings and 200x speed improvements over frontier models for structured tasks like summarization and extraction.
The open-source harness moves context out of the active window using offloading, summarization, skills, subagents, and long-term memory.
Managed Deep Agents v0.9 introduces a new reactions attribute for Slack channels, allowing developers to dynamically assign emoji responses using simple heuristics or decision models like Jev.
AI Briefing