이 요약은 AI가 원문을 분석해 생성했습니다. 정확한 내용은 원문 기준으로 확인하세요.
TL;DR
11x의 CTO Jeson Patel은 기존에 PM의 개인 PC에서 수동으로 처리되던 버그 트리아지 프로세스를 LangSmith Fleet을 활용해 Slack 기반의 AI 에이전트로 전환한 사례를 공유한다. 인프라 구축 없이 에이전트를 배포하고 관리할 수 있는 Fleet의 편의성을 바탕으로, 버그 대응뿐만 아니라 일반적인 Q&A와 영업 지원 등 다양한 사내 업무로 에이전트 활용 범위를 확장했다. 그는 이러한 도구의 접근성이 조직 내 모든 구성원이 각자의 업무 워크플로를 자동화하는 에이전트를 직접 구축할 수 있는 환경을 조성한다고 강조한다.
챕터별 상세
00:00
Meet the CTO
Jeson Patel, CTO at 11x, introduces the company's mission to build AI software for go-to-market teams. He explains his role in overseeing the technical strategy and the integration of AI into their internal processes. This sets the stage for discussing the practical application of AI agents in a business environment. He emphasizes the importance of building tools that directly address the needs of go-to-market teams.
00:09
The bug triage problem
The team faced a challenge where bug triaging was manual and tied to a single PM's machine, creating a bottleneck. This reliance on a single person's local setup was inefficient and not scalable. The team needed a more robust and automated solution to handle the increasing volume of bug reports. They realized that manual processes were hindering their ability to respond quickly to issues.
00:22
Infrastructure-free agents
The goal was to build AI agents that could handle bug triage without the burden of managing complex infrastructure code. Jeson sought a solution that would allow the team to focus on agent logic rather than infrastructure maintenance. This requirement was driven by the need for speed and agility in their development process. They wanted to avoid the overhead of building and maintaining custom infrastructure.
00:35
Finding Fleet
Jeson discovered LangSmith Fleet and found its capabilities for building agents to be highly effective and efficient. He initially thought the tool's ease of use was too good to be true, but it proved to be a powerful solution for their needs. The platform allowed them to build and deploy agents without writing extensive infrastructure code. This enabled the team to rapidly prototype and iterate on their agent designs.
00:42
Slack integration
The agent was integrated directly into Slack, allowing team members to easily interact with it by tagging it in channels. This Slack-native approach ensured that the agent was accessible to everyone in the company without requiring a separate interface. The team simply tags the agent to trigger its bug triage and investigation capabilities. This seamless integration reduced friction and increased the adoption of the agent among team members.
00:52
General-purpose agent
The agent's scope expanded from simple bug triage to handling general Q&A and other internal inquiries. This expansion demonstrated the versatility of the agent and its ability to adapt to different tasks. The team realized that the same underlying agent architecture could be applied to various business problems. This flexibility allowed them to address multiple pain points with a single, unified solution.
01:06
Folding use cases
Multiple use cases were consolidated into a single agent, increasing its utility and capabilities over time. By folding different tasks into one agent, the team reduced complexity and improved the overall user experience. This consolidation allowed the agent to become a central hub for internal information and support. It also simplified the management and maintenance of their AI agents.
01:24
Sales usage
Salespeople began using the agent during calls to quickly answer product-related questions. This real-time access to information improved the efficiency of sales calls and provided immediate value to the team. The agent's ability to handle diverse queries made it an indispensable tool for the sales department. Salespeople could now provide accurate and timely information to prospects without needing to consult other team members.
01:35
Org-wide adoption
The next phase involves enabling the entire organization to build and deploy their own agents. Jeson envisions a future where every team member can create agents tailored to their specific workflows. This democratization of agent building is expected to drive further innovation and efficiency across the company. He believes that empowering employees to build their own tools will lead to more effective and relevant solutions.
01:44
Democratizing agent building
The ease of use of Fleet empowers every team member to automate their own workflows, scaling the impact of AI. By lowering the barrier to entry, the company can leverage the collective expertise of its employees to build more effective AI solutions. This approach transforms AI from a specialized tool into a pervasive asset for the entire organization. It enables a culture of continuous improvement and automation.
용어 해설
- AI Agent
- — LLM을 기반으로 도구를 사용하고 추론하여 특정 작업을 자율적으로 수행하는 시스템입니다. 이 아티클에서는 버그 트리아지나 Q&A와 같은 사내 업무를 자동화하는 데 활용됩니다.
- Bug Triage
- — 소프트웨어 개발 과정에서 보고된 버그를 분석하고 우선순위를 정해 담당자에게 배정하는 과정입니다. 수동 프로세스를 자동화하여 효율성을 높이는 것이 중요합니다.
- Slack Integration
- — 외부 서비스나 AI 에이전트를 슬랙 채널 내에서 직접 호출하고 대화할 수 있게 연결하는 방식입니다. 팀원들이 익숙한 도구에서 AI 기능을 즉시 활용할 수 있게 합니다.
언급된 리소스
AI 분석 전체 내용 보기
AI 요약 · 북마크 · 개인 피드 설정 — 무료
출처 · 인용 안내
원문 발행 2026. 07. 16.수집 2026. 07. 16.출처 타입 YOUTUBE
인용 시 "요약 출처: AI Trends (aitrends.kr)"를 표기하고, 사실 확인은 원문 보기 기준으로 진행해 주세요. 자세한 기준은 운영 정책을 참고해 주세요.
