{"id":"79f85f13100be6761c8488aac1da55d381690c900de0c25e9135d0badfd078ef","title":"OpenAI’s Performance Factory automates performance-regression work","argument":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes describe the Performance Factory, where Codex agents investigate performance alerts, propose fixes, and check whether performance recovers after deployment. The authors report that its Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months. Engineers still review and deploy proposed fixes; agents check the result and can send unresolved slowdowns back for investigation.\n\nThe authors trace a shift from manual investigations, to engineers working with Codex, to reusable skills and automated workflows. They say agents sometimes settle on plausible but wrong diagnoses, so a separate Codex session challenges proposed causes. In a manual review of 100 root cause suggestions, they graded 70% as accurate; 30% needed refinement or did not fully establish causality. Fixes can be more complicated than necessary or introduce new issues, and some require engineers to weigh tradeoffs.\n\nThe authors evaluate investigations against past incidents and assess whether task summaries help engineers understand findings. They are working on synthetic performance tests to measure proposed fixes before engineer review, and on finding slowdowns that alerts miss. They are also exploring whether agents could eventually ship some low-risk fixes without a human in the loop.","topicIds":["intelligence-and-discovery","work-and-agency"],"domain":"work","claimType":"attributed reading","limits":"The reported accuracy figure comes from a manual review of 100 root cause suggestions; the article does not present it as a measure of overall real-world reliability. The authors describe synthetic performance tests, proactive monitoring, and possible agent-led shipping of low-risk fixes as work in progress, not demonstrated outcomes.","speaker":{"name":"OpenAI","roleAtSource":"Publisher"},"source":{"title":"How OpenAI’s Performance Factory keeps ChatGPT fast","publisher":"OpenAI","format":"Article","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","publishedAt":"2026-11-19T08:00:00.000Z","dateBasis":"Date supplied by the publisher’s feed."},"evidence":[{"text":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes say the Performance Factory automates performance-regression work, from investigating alerts to proposing fixes and checking recovery after deployment.","excerpt":"The Performance Factory, which runs in the cloud and automates investigating and resolving performance problems","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors report that the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months.","excerpt":"In just over three months, the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors added a separate Codex session to challenge proposed diagnoses after some agents stopped at plausible explanations that proved wrong.","excerpt":"To catch these mistakes, we added an adversarial review: a separate Codex session that tries to disprove the proposed diagnosis.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"In a manual review of 100 root cause suggestions, the authors graded 70% accurate, while 30% needed refinement or did not fully establish causality.","excerpt":"In a manual review of 100 root cause suggestions, we graded 70% as accurate, with 30% needing further refinement or not fully establishing causality.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors say proposed fixes can be unnecessarily complicated or introduce new issues; engineers review and decide whether to merge them.","excerpt":"we’ve seen cases where the agent-proposed changes are more complicated than necessary or introduce new issues.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors are working on synthetic performance tests to measure proposed fixes and revise them before engineer review.","excerpt":"We’re also working on using synthetic performance tests so agents can measure proposed fixes and revise them before asking an engineer for review.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"}],"sourceMedia":{"status":"found","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","checkedAt":"2026-11-21T17:45:56.904Z","items":[{"type":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","attribution":"OpenAI","basis":"publisher-embed","availability":"verified","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex, engineers dispatching tasks, systems dispatching tasks, an…","verifiedAt":"2026-11-21T17:45:57.536Z"}],"note":"Media declared by the original publisher; availability checked separately from the editorial reading."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What results will the synthetic performance tests produce when used to measure and revise agent-proposed fixes before engineer review?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"OpenAI’s Performance Factory automates performance-regression work","summary":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes describe the Performance Factory, where Codex agents investigate performance alerts, propose fixes, and check whether performance recovers after deployment. The authors report that its Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months. Engineers still review and deploy proposed fixes; agents check the result and can send unresolved slowdowns back for investigation.\n\nThe authors trace a shift from manual investigations, to engineers working with Codex, to reusable skills and automated workflows. They say agents sometimes settle on plausible but wrong diagnoses, so a separate Codex session challenges proposed causes. In a manual review of 100 root cause suggestions, they graded 70% as accurate; 30% needed refinement or did not fully establish causality. Fixes can be more complicated than necessary or introduce new issues, and some require engineers to weigh tradeoffs.\n\nThe authors evaluate investigations against past incidents and assess whether task summaries help engineers understand findings. They are working on synthetic performance tests to measure proposed fixes before engineer review, and on finding slowdowns that alerts miss. They are also exploring whether agents could eventually ship some low-risk fixes without a human in the loop.","question":"What results will the synthetic performance tests produce when used to measure and revise agent-proposed fixes before engineer review?","claims":[{"text":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes say the Performance Factory automates performance-regression work, from investigating alerts to proposing fixes and checking recovery after deployment.","excerpt":"The Performance Factory, which runs in the cloud and automates investigating and resolving performance problems","support":"source-stated"},{"text":"The authors report that the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months.","excerpt":"In just over three months, the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases.","support":"source-stated"},{"text":"The authors added a separate Codex session to challenge proposed diagnoses after some agents stopped at plausible explanations that proved wrong.","excerpt":"To catch these mistakes, we added an adversarial review: a separate Codex session that tries to disprove the proposed diagnosis.","support":"source-stated"},{"text":"In a manual review of 100 root cause suggestions, the authors graded 70% accurate, while 30% needed refinement or did not fully establish causality.","excerpt":"In a manual review of 100 root cause suggestions, we graded 70% as accurate, with 30% needing further refinement or not fully establishing causality.","support":"source-stated"},{"text":"The authors say proposed fixes can be unnecessarily complicated or introduce new issues; engineers review and decide whether to merge them.","excerpt":"we’ve seen cases where the agent-proposed changes are more complicated than necessary or introduce new issues.","support":"source-stated"},{"text":"The authors are working on synthetic performance tests to measure proposed fixes and revise them before engineer review.","excerpt":"We’re also working on using synthetic performance tests so agents can measure proposed fixes and revise them before asking an engineer for review.","support":"source-stated"}],"limitations":["The reported accuracy figure comes from a manual review of 100 root cause suggestions; the article does not present it as a measure of overall real-world reliability.","The authors describe synthetic performance tests, proactive monitoring, and possible agent-led shipping of low-risk fixes as work in progress, not demonstrated outcomes."],"mediaId":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","correctionReason":null,"id":"79f85f13100be6761c8488aac1da55d381690c900de0c25e9135d0badfd078ef","publicationId":"10095054-8779-4e72-87a3-fbb90bb3d54f","version":1,"sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","sourceTitle":"How OpenAI’s Performance Factory keeps ChatGPT fast","publisher":"OpenAI","publishedAt":"2026-11-19T08:00:00.000Z","originalPublished":{"value":"2026-11-19T08:00:00.000Z","precision":"instant","basis":"feed_entry"},"firstObservedAt":"2026-11-19T17:43:01.492Z","createdAt":"2026-11-21T17:46:59.025Z","historical":false,"entityIds":["openai"],"topicIds":["intelligence-and-discovery","work-and-agency"],"media":[{"id":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","kind":"article","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-21T17:45:57.536Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."},{"id":"11a707d2f2b324d4798344eebb5dcc814af3c97191c052d54cef1e2f64ae0b2e","kind":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex,","provenance":"publisher_page","verification":{"status":"verified","checkedAt":"2026-11-21T17:45:57.536Z","method":"publisher_association_and_head","contentType":"image/webp","cacheUntil":1795283457536},"recommendationReason":"Observed on the acquired publisher source; availability checked."}],"sourceMedia":{"status":"found","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","checkedAt":"2026-11-21T17:45:56.904Z","items":[{"type":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","attribution":"OpenAI","basis":"publisher-embed","availability":"verified","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex, engineers dispatching tasks, systems dispatching tasks, an…","verifiedAt":"2026-11-21T17:45:57.536Z"}],"note":"Media declared by the original publisher; availability checked separately from the editorial reading."},"editorialDecision":{"outcome":"accept","reasonCodes":["useful_perspective"],"rubricVersion":"stream-admission-v9","mediaId":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","confidence":0.8400000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"28965dc999118695405f4b1c629e4ab28ca46c001bf5c86be46aa32249ac6b15","decidedAt":"2026-11-21T17:45:59.982Z"},"provenance":{"textHash":"2e5b367c12f5cf12343d510200b15d610a59f4bf9e155b00218d6a5d3d75da55","observationId":"c3b2e9af66f55a659066239466a753f5d58992d349eac4a254903084f2c334cc","retrievedAt":"2026-11-21T17:45:56.904Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"18fe8ec2-6abe-40fb-8632-71a82e4f483a"},"reviewer":{"model":"gpt-6-luna","runId":"2bd56f21-1e56-48ea-8d87-2b728e2faeb9"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-21T17:46:59.025Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":null},"canonicalUrl":"https://escapemeridian.com/readings/openai-s-performance-factory-automates-performance-regression-79f85f13","formats":{"html":"/readings/openai-s-performance-factory-automates-performance-regression-79f85f13","markdown":"/readings/openai-s-performance-factory-automates-performance-regression-79f85f13.md","json":"/readings/openai-s-performance-factory-automates-performance-regression-79f85f13.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"conditions":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[],"entities":[{"id":"openai","name":"OpenAI","kind":"organization","context":"OpenAI develops AI models and products and publishes research, release announcements and evaluations.","referenceUrls":["https://openai.com/news/"],"path":"/entities/openai","json":"/entities/openai.json","markdown":"/entities/openai.md"},{"id":"figure","name":"Figure","kind":"organization","context":"Figure develops humanoid robots and publishes its own tests of how they perform physical tasks.","referenceUrls":["https://www.figure.ai/news","https://www.figure.ai/news/helix-2-5-zero-shot-30-home-generalization"],"path":"/entities/figure","json":"/entities/figure.json","markdown":"/entities/figure.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"79f85f13100be6761c8488aac1da55d381690c900de0c25e9135d0badfd078ef","revision":"10095054-8779-4e72-87a3-fbb90bb3d54f","revisionKind":"publication-id","title":"OpenAI’s Performance Factory automates performance-regression work","url":"https://escapemeridian.com/readings/openai-s-performance-factory-automates-performance-regression-79f85f13"},"source":{"type":"source","id":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","revision":"2e5b367c12f5cf12343d510200b15d610a59f4bf9e155b00218d6a5d3d75da55","revisionKind":"sha256-source-text","title":"How OpenAI’s Performance Factory keeps ChatGPT fast","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings to this reading. Shared topics are not citations. Other record types need an explicit reference adapter."},"revisions":[{"id":"79f85f13100be6761c8488aac1da55d381690c900de0c25e9135d0badfd078ef","title":"OpenAI’s Performance Factory automates performance-regression work","argument":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes describe the Performance Factory, where Codex agents investigate performance alerts, propose fixes, and check whether performance recovers after deployment. The authors report that its Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months. Engineers still review and deploy proposed fixes; agents check the result and can send unresolved slowdowns back for investigation.\n\nThe authors trace a shift from manual investigations, to engineers working with Codex, to reusable skills and automated workflows. They say agents sometimes settle on plausible but wrong diagnoses, so a separate Codex session challenges proposed causes. In a manual review of 100 root cause suggestions, they graded 70% as accurate; 30% needed refinement or did not fully establish causality. Fixes can be more complicated than necessary or introduce new issues, and some require engineers to weigh tradeoffs.\n\nThe authors evaluate investigations against past incidents and assess whether task summaries help engineers understand findings. They are working on synthetic performance tests to measure proposed fixes before engineer review, and on finding slowdowns that alerts miss. They are also exploring whether agents could eventually ship some low-risk fixes without a human in the loop.","topicIds":["intelligence-and-discovery","work-and-agency"],"domain":"work","claimType":"attributed reading","limits":"The reported accuracy figure comes from a manual review of 100 root cause suggestions; the article does not present it as a measure of overall real-world reliability. The authors describe synthetic performance tests, proactive monitoring, and possible agent-led shipping of low-risk fixes as work in progress, not demonstrated outcomes.","speaker":{"name":"OpenAI","roleAtSource":"Publisher"},"source":{"title":"How OpenAI’s Performance Factory keeps ChatGPT fast","publisher":"OpenAI","format":"Article","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","publishedAt":"2026-11-19T08:00:00.000Z","dateBasis":"Date supplied by the publisher’s feed."},"evidence":[{"text":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes say the Performance Factory automates performance-regression work, from investigating alerts to proposing fixes and checking recovery after deployment.","excerpt":"The Performance Factory, which runs in the cloud and automates investigating and resolving performance problems","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors report that the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months.","excerpt":"In just over three months, the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors added a separate Codex session to challenge proposed diagnoses after some agents stopped at plausible explanations that proved wrong.","excerpt":"To catch these mistakes, we added an adversarial review: a separate Codex session that tries to disprove the proposed diagnosis.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"In a manual review of 100 root cause suggestions, the authors graded 70% accurate, while 30% needed refinement or did not fully establish causality.","excerpt":"In a manual review of 100 root cause suggestions, we graded 70% as accurate, with 30% needing further refinement or not fully establishing causality.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors say proposed fixes can be unnecessarily complicated or introduce new issues; engineers review and decide whether to merge them.","excerpt":"we’ve seen cases where the agent-proposed changes are more complicated than necessary or introduce new issues.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"},{"text":"The authors are working on synthetic performance tests to measure proposed fixes and revise them before engineer review.","excerpt":"We’re also working on using synthetic performance tests so agents can measure proposed fixes and revise them before asking an engineer for review.","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","relationship":"Source-stated"}],"sourceMedia":{"status":"found","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","checkedAt":"2026-11-21T17:45:56.904Z","items":[{"type":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","attribution":"OpenAI","basis":"publisher-embed","availability":"verified","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex, engineers dispatching tasks, systems dispatching tasks, an…","verifiedAt":"2026-11-21T17:45:57.536Z"}],"note":"Media declared by the original publisher; availability checked separately from the editorial reading."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What results will the synthetic performance tests produce when used to measure and revise agent-proposed fixes before engineer review?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"OpenAI’s Performance Factory automates performance-regression work","summary":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes describe the Performance Factory, where Codex agents investigate performance alerts, propose fixes, and check whether performance recovers after deployment. The authors report that its Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months. Engineers still review and deploy proposed fixes; agents check the result and can send unresolved slowdowns back for investigation.\n\nThe authors trace a shift from manual investigations, to engineers working with Codex, to reusable skills and automated workflows. They say agents sometimes settle on plausible but wrong diagnoses, so a separate Codex session challenges proposed causes. In a manual review of 100 root cause suggestions, they graded 70% as accurate; 30% needed refinement or did not fully establish causality. Fixes can be more complicated than necessary or introduce new issues, and some require engineers to weigh tradeoffs.\n\nThe authors evaluate investigations against past incidents and assess whether task summaries help engineers understand findings. They are working on synthetic performance tests to measure proposed fixes before engineer review, and on finding slowdowns that alerts miss. They are also exploring whether agents could eventually ship some low-risk fixes without a human in the loop.","question":"What results will the synthetic performance tests produce when used to measure and revise agent-proposed fixes before engineer review?","claims":[{"text":"OpenAI Technical Staff members Okay Zed and Cyrus Forbes say the Performance Factory automates performance-regression work, from investigating alerts to proposing fixes and checking recovery after deployment.","excerpt":"The Performance Factory, which runs in the cloud and automates investigating and resolving performance problems","support":"source-stated"},{"text":"The authors report that the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases in just over three months.","excerpt":"In just over three months, the Regression Workflow tracked more than 1,700 investigations and root caused over 600 cases.","support":"source-stated"},{"text":"The authors added a separate Codex session to challenge proposed diagnoses after some agents stopped at plausible explanations that proved wrong.","excerpt":"To catch these mistakes, we added an adversarial review: a separate Codex session that tries to disprove the proposed diagnosis.","support":"source-stated"},{"text":"In a manual review of 100 root cause suggestions, the authors graded 70% accurate, while 30% needed refinement or did not fully establish causality.","excerpt":"In a manual review of 100 root cause suggestions, we graded 70% as accurate, with 30% needing further refinement or not fully establishing causality.","support":"source-stated"},{"text":"The authors say proposed fixes can be unnecessarily complicated or introduce new issues; engineers review and decide whether to merge them.","excerpt":"we’ve seen cases where the agent-proposed changes are more complicated than necessary or introduce new issues.","support":"source-stated"},{"text":"The authors are working on synthetic performance tests to measure proposed fixes and revise them before engineer review.","excerpt":"We’re also working on using synthetic performance tests so agents can measure proposed fixes and revise them before asking an engineer for review.","support":"source-stated"}],"limitations":["The reported accuracy figure comes from a manual review of 100 root cause suggestions; the article does not present it as a measure of overall real-world reliability.","The authors describe synthetic performance tests, proactive monitoring, and possible agent-led shipping of low-risk fixes as work in progress, not demonstrated outcomes."],"mediaId":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","correctionReason":null,"id":"79f85f13100be6761c8488aac1da55d381690c900de0c25e9135d0badfd078ef","publicationId":"10095054-8779-4e72-87a3-fbb90bb3d54f","version":1,"sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","sourceTitle":"How OpenAI’s Performance Factory keeps ChatGPT fast","publisher":"OpenAI","publishedAt":"2026-11-19T08:00:00.000Z","originalPublished":{"value":"2026-11-19T08:00:00.000Z","precision":"instant","basis":"feed_entry"},"firstObservedAt":"2026-11-19T17:43:01.492Z","createdAt":"2026-11-21T17:46:59.025Z","historical":false,"entityIds":["openai"],"topicIds":["intelligence-and-discovery","work-and-agency"],"media":[{"id":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","kind":"article","url":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-21T17:45:57.536Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."},{"id":"11a707d2f2b324d4798344eebb5dcc814af3c97191c052d54cef1e2f64ae0b2e","kind":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex,","provenance":"publisher_page","verification":{"status":"verified","checkedAt":"2026-11-21T17:45:57.536Z","method":"publisher_association_and_head","contentType":"image/webp","cacheUntil":1795283457536},"recommendationReason":"Observed on the acquired publisher source; availability checked."}],"sourceMedia":{"status":"found","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","checkedAt":"2026-11-21T17:45:56.904Z","items":[{"type":"image","url":"https://images.ctfassets.net/kftzwdyauwt9/581rKVIvguLraKii21zIgr/83deca5cdcec639b656a9a5ec1954f4b/performance-factory-evolution-light.png?w=1200&q=90&fm=webp","sourceUrl":"https://openai.com/index/how-openai-performance-factory-keeps-chatgpt-fast/","attribution":"OpenAI","basis":"publisher-embed","availability":"verified","label":"The evolution of performance engineering: manual work, reusable tools, engineers working with Codex, engineers dispatching tasks, systems dispatching tasks, an…","verifiedAt":"2026-11-21T17:45:57.536Z"}],"note":"Media declared by the original publisher; availability checked separately from the editorial reading."},"editorialDecision":{"outcome":"accept","reasonCodes":["useful_perspective"],"rubricVersion":"stream-admission-v9","mediaId":"9342dbf934e428aa00c16520441f258a949ccd62bcb28471deb475c04ce3c227","confidence":0.8400000000000001,"provider":"typesafe","model":"jev-1.13.0","inputHash":"28965dc999118695405f4b1c629e4ab28ca46c001bf5c86be46aa32249ac6b15","decidedAt":"2026-11-21T17:45:59.982Z"},"provenance":{"textHash":"2e5b367c12f5cf12343d510200b15d610a59f4bf9e155b00218d6a5d3d75da55","observationId":"c3b2e9af66f55a659066239466a753f5d58992d349eac4a254903084f2c334cc","retrievedAt":"2026-11-21T17:45:56.904Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"18fe8ec2-6abe-40fb-8632-71a82e4f483a"},"reviewer":{"model":"gpt-6-luna","runId":"2bd56f21-1e56-48ea-8d87-2b728e2faeb9"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-21T17:46:59.025Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":null}}],"audits":[{"publicationId":"10095054-8779-4e72-87a3-fbb90bb3d54f","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The draft correctly attributes the article to OpenAI's Okay Zed and Cyrus Forbes, Members of Technical Staff. It keeps the reported figures and plans attributed to the authors rather than presenting them as established facts.","excerpt":"By Okay Zed and Cyrus Forbes, Members of Technical Staff"},{"dimension":"summary","status":"uncertain","explanation":"The mechanics are accurate, but the reading leaves out the authors' central argument for why this matters. They say agent-driven development now moves so fast that traditional performance work cannot keep up, so performance systems must become autonomous. Without that, the summary reads as a feature list. Smallest repair: add one attributed sentence stating that thesis. Also gloss Codex in plain words as OpenAI's coding agent, and regression as a slowdown after a change.","excerpt":"The staggering pace of agentic development has made traditional approaches to performance engineering unsustainable."},{"dimension":"date","status":"pass","explanation":"The originalPublished value, 2026-11-19, matches the article's own dateline. No earlier publication is indicated.","excerpt":"November 19, 2026"},{"dimension":"support","status":"pass","explanation":"Every claim matches a verbatim source passage: the 1,700 investigations and 600 root causes, the adversarial review, the 70% figure, the fix problems, engineer review, synthetic tests and low-risk shipping. The limitations correctly frame the 70% figure and future plans. The question follows from stated work in progress without asserting an outcome.","excerpt":"We’re also exploring how agents could use our other agentic review tools to identify low-risk changes and eventually ship those fixes without a human in the loop."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct link to the original OpenAI article, which is the most direct way to access the source.","excerpt":null}],"createdAt":"2026-11-21T17:46:59.025Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}