影片進階EN1.6 萬 次觀看
The State of AI in Software Development: Data from 400+ Orgs — Justin Reock, DX
來源 AI Engineer
看影片(在新分頁開啟原站)連到 AI Engineer
摘要
DX 透過 400 多家組織的資料,發現 AI 雖提升部署頻率,但變更失敗率波動劇增,可維護性上升而信任度下降。 Junior 工程師使用 AI 最多,但 Staff+ 工程師能節省更多時間且消耗更少 Token。AI 測量框架強調利用、影響與成本,並透過 Agent 反饋評估平台準備度。案例顯示程式碼生成從未是瓶頸,Morgan Stanley 年省 30 萬小時,Zapier 每工程師價值提升 15%,Spotify 則透過 AI 事故代理大幅縮短修復流程。
DX analyzed data from 400+ organizations, revealing that while AI boosts deployment frequency, change failure rates become more volatile, maintainability rises while trust drops. Junior engineers use AI most, but Staff+ engineers save more time and consume fewer tokens. The AI measurement framework…
重點
- 部署頻率上升,但變更失敗率波動劇增,可維護性提高而信任度下降。
- Junior 工程師使用 AI 最多,但 Staff+ 工程師能節省更多時間且消耗更少 Token。
- AI 測量框架強調利用、影響與成本,並透過 Agent 反饋評估平台準備度。
章節
依話題轉折切分,標題由 AI 產生
- 00:00Intro: DX research on AI's impact
- 01:27Trend 1: velocity and deployment frequency
- 02:53Regional differences
- 03:33Perceived speed vs. reality
- 04:38Trend 2: quality and change failure rate volatility
- 06:02Maintainability up, change confidence down
- 07:02PR size is growing and incremental delivery is suffering
- 08:42Trend 3: juniors, staff+ engineers and company size
- 10:37Trend 4: how to measure AI (utilization, impact, cost)
- 13:27Platform AI readiness: good DX is good agent experience
- 14:17Listening to feedback from agents
- 14:57Trend 5: code generation was never the bottleneck
- 16:27Case studies: Morgan Stanley, Zapier, Spotify
- 18:32Get the Q2 report
提到的工具與公司
- DevX framework
- DevGen AI agent
- Zapier
- PRD
- SR communication channels
適合誰看
軟體工程師、DevOps 團隊、希望量化 AI 對效能影響的組織管理者。
摘要依據
- 講者
- Justin Reock
- 依據
- 自動字幕
為什麼排在這裡
- 人氣
- 0.83
- 新鮮
- 0.99
在主題頁與搜尋結果裡,名次由相關、人氣、新鮮三個分數決定;這一頁沒有搜尋的關鍵字,所以沒有相關分數。排序怎麼算
相關內容
- Measuring the impact of AI on software engineering – with Laura TachoPodcast ・ The Pragmatic Engineer ・ 1 小時 12 分(在新分頁開啟原站)
- AI 时代到底该怎么管一个工程团队文章 ・ 寶玉
- Agents, IDEs, and the Blast Radius: Practical AI for Software EngineersPodcast ・ AI Engineering Podcast ・ 59 分鐘
- Stop being skeptical about AI for development with Charity MajorsPodcast ・ The Pragmatic Engineer ・ 1 小時 26 分
- AI's 70% problem w/ Addy Osmani影片 ・ Zed Industries ・ 48 分鐘(在新分頁開啟原站)
- Why Agents Are Forcing Enterprises to Finally Fix Their Dev ProcessPodcast ・ The AI Native Dev ・ 1 小時
摘要由 AI 根據原文產生,可能有誤;完整內容請看原站。看影片(在新分頁開啟原站)
