Lenny Rachitsky · Oct 19, 2026
Lenny’s Newsletter: Meta Muse Wonder and Brooke Lacey’s Claude video workflow
Markdown · JSON · Original source
Lenny’s Newsletter presents two accounts of AI-assisted work: Claire’s tests of Meta Muse Wonder and creator Brooke Lacey’s video-editing workflow. Claire reports that Muse Wonder handled tasks such as billing support and inbox triage well in her tests, but struggled with design and finishing prototypes. She plans to try it in some agent workflows, not use it for frontend design. Lacey describes using Code, FFmpeg, and Whisper to edit phone recordings into captioned videos. She says she taught Claude her jump-cut cadence through instruction and iteration, and uses audience comments to develop video ideas. The workflow removes editing friction, she says, but her experience with video craft is what lets her judge whether the edits work. Escape Meridian interpretation: Together, the accounts distinguish delegating execution from supplying judgment: Claire’s results vary by task, while Lacey says knowing the craft mattered to teaching the system her style. The descriptions do not establish how these workflows perform beyond the reported tests and examples. What results would Claire’s agent workflows produce outside her tests, and how well would Lacey’s editing process transfer to other creators’ styles?
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.
Press play to watch. If playback is unavailable, open the source below.
Original source
· Article
Source details
Original source date: Oct 19, 2026 (2026-10-19T15:03:35+00:00).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-05T04:08:14.195Z. First observed: 2026-11-04T12:36:38.419Z. Retrieved: 2026-11-05T03:55:35.157Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The source is a newsletter description of two podcast episodes, not a transcript; the details and conclusions are presented as the speakers’ accounts. Claire is identified only by first name in the supplied text, so her identity and the methods behind her tests are not established here. The source does not report results from Claire’s planned agent workflows or establish how either workflow performs beyond the described examples.
Open question
What results would Claire’s agent workflows produce outside her tests, and how well would Lacey’s editing process transfer to other creators’ styles?
Source-stated
Claire’s tests found Muse Wonder strong on several agentic tasks and weaker on design and prototype finishing.
It ranked first on billing-ticket resolution, inbox triage, computer use, and email writing.
Source-stated
Claire plans to test Muse Wonder in some agent workflows, while avoiding frontend design work.
She plans to test it inside a few agent workflows because its multi-step reasoning and natural voice are strong.
Source-stated
Brooke Lacey says Claude Code, FFmpeg, and Whisper let her turn raw phone video into an edited, captioned video quickly.
A raw phone video goes in, and about 15 seconds later, a finished vertical short comes out with captions, jump cuts, and zooms.
Source-stated
Lacey says teaching Claude her editing style took deliberate instruction and iteration, and depended on her own knowledge of video craft.
Encoding that style required deliberate instruction and iteration, not one clever prompt.
Escape Meridian interpretation
The two accounts suggest that reported benefits depend on the task and on human judgment; the descriptions do not establish broader performance.
AI amplifies that judgment, but it does not replace the experience required to develop it.
Source and contributor context
Lenny Rachitsky · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Agentic AI
AI systems that plan and carry out a series of steps toward a goal, often by using other software, rather than giving one answer.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Lenny's Newsletter
Lenny's Newsletter is Lenny Rachitsky's publication and podcast on product management, growth and career questions for product people.
Editorial provenance
Version 1. Publication ID: 6a2137ec-9214-4b1a-899c-8316af94f6c1.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-05T04:08:14.195Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: useful_perspective.
Observation: f1c702834abe027854bdcd3aa890e24c72283ec00d4a95f7c8b6e632432ac10a. Source text hash: bc0d865a8aa4455990873abf4d623b20cb90350d54389774c85df4a3173540a1.
Revision and source-check history
Version 1 · 2026-11-05T04:08:14.195Z
Lenny’s Newsletter: Meta Muse Wonder and Brooke Lacey’s Claude video workflow
Lenny’s Newsletter presents two accounts of AI-assisted work: Claire’s tests of Meta Muse Wonder and creator Brooke Lacey’s video-editing workflow. Claire reports that Muse Wonder handled tasks such as billing support and inbox triage well in her tests, but struggled with design and finishing prototypes. She plans to try it in some agent workflows, not use it for frontend design. Lacey describes using Code, FFmpeg, and Whisper to edit phone recordings into captioned videos. She says she taught Claude her jump-cut cadence through instruction and iteration, and uses audience comments to develop video ideas. The workflow removes editing friction, she says, but her experience with video craft is what lets her judge whether the edits work. Escape Meridian interpretation: Together, the accounts distinguish delegating execution from supplying judgment: Claire’s results vary by task, while Lacey says knowing the craft mattered to teaching the system her style. The descriptions do not establish how these workflows perform beyond the reported tests and examples. What results would Claire’s agent workflows produce outside her tests, and how well would Lacey’s editing process transfer to other creators’ styles?
Original source
· Article
Source details
Original source date: Oct 19, 2026 (2026-10-19T15:03:35+00:00).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-05T04:08:14.195Z. First observed: 2026-11-04T12:36:38.419Z. Retrieved: 2026-11-05T03:55:35.157Z.
Limits
The source is a newsletter description of two podcast episodes, not a transcript; the details and conclusions are presented as the speakers’ accounts. Claire is identified only by first name in the supplied text, so her identity and the methods behind her tests are not established here. The source does not report results from Claire’s planned agent workflows or establish how either workflow performs beyond the described examples.
Open question
What results would Claire’s agent workflows produce outside her tests, and how well would Lacey’s editing process transfer to other creators’ styles?
Source-stated
Claire’s tests found Muse Wonder strong on several agentic tasks and weaker on design and prototype finishing.
It ranked first on billing-ticket resolution, inbox triage, computer use, and email writing.
Source-stated
Claire plans to test Muse Wonder in some agent workflows, while avoiding frontend design work.
She plans to test it inside a few agent workflows because its multi-step reasoning and natural voice are strong.
Source-stated
Brooke Lacey says Claude Code, FFmpeg, and Whisper let her turn raw phone video into an edited, captioned video quickly.
A raw phone video goes in, and about 15 seconds later, a finished vertical short comes out with captions, jump cuts, and zooms.
Source-stated
Lacey says teaching Claude her editing style took deliberate instruction and iteration, and depended on her own knowledge of video craft.
Encoding that style required deliberate instruction and iteration, not one clever prompt.
Escape Meridian interpretation
The two accounts suggest that reported benefits depend on the task and on human judgment; the descriptions do not establish broader performance.
AI amplifies that judgment, but it does not replace the experience required to develop it.
Source and contributor context
Lenny Rachitsky · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Agentic AI
AI systems that plan and carry out a series of steps toward a goal, often by using other software, rather than giving one answer.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Lenny's Newsletter
Lenny's Newsletter is Lenny Rachitsky's publication and podcast on product management, growth and career questions for product people.
Editorial provenance
Version 1. Publication ID: 6a2137ec-9214-4b1a-899c-8316af94f6c1.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-05T04:08:14.195Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: useful_perspective.
Observation: f1c702834abe027854bdcd3aa890e24c72283ec00d4a95f7c8b6e632432ac10a. Source text hash: bc0d865a8aa4455990873abf4d623b20cb90350d54389774c85df4a3173540a1.
passed · 2026-11-05T04:08:14.195Z · anthropic/claude-opus-5.5
Publication: 6a2137ec-9214-4b1a-899c-8316af94f6c1
attribution — pass: The draft attributes the piece to Lenny's Newsletter, the Muse Wonder findings to Claire's tests, and the workflow claims to Brooke Lacey. It does not present their claims as established fact. The limitation correctly notes that Claire is identified only by first name in the supplied text.
Claire puts Meta’s new Muse Wonder model through nearly 20 real-world tasks covering agentic work, coding, writing, design, and video editing.
summary — pass: The title, summary and question faithfully reflect both segments. They cover Muse Wonder's agentic strengths, its design and prototype weaknesses, and Claire's limited adoption plan. They also cover Lacey's workflow, how she taught Claude her style, her use of comments, and her emphasis on craft knowledge. Technical terms such as agentic, FFmpeg, Whisper and Claude Code are kept exactly as the source uses them, which is appropriate for term-library tooltips. The interpretation is labeled as Escape Meridian's and stays conditional. The question addresses a real open issue the source leaves unresolved.
Creators need to understand the craft before they can delegate it.
date — pass: The originalPublished timestamp, 2026-10-19, matches the publisher page dateline (Oct 19, 2026) and the feed entry.
Oct 19, 2026
support — pass: Each source-stated claim is backed by text in the source: Muse Wonder ranking first on agentic tasks, its design failures, Claire's planned limited testing, Lacey's roughly 15-second workflow, and her instruction-and-iteration process. The interpretive claim is labeled as such and grounded in Lacey's statement about judgment. The draft also avoids extending the reported tests into general reliability.
Her Claude Code workflow didn’t help her post more; it removed hours of editing friction so each video no longer consumes an entire day.
media — pass: The selected media is the verified direct link to the original newsletter article, which covers both episodes. That makes it an appropriate and relevant way for readers to access the source.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
AI systems that plan and carry out a series of steps toward a goal, often by using other software, rather than giving one answer.
Definition referenceClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeLenny's Newsletter is Lenny Rachitsky's publication and podcast on product management, growth and career questions for product people.
Explore Lenny's NewsletterAI systems that plan and carry out a series of steps toward a goal, often by using other software, rather than giving one answer.
Definition referenceClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeLenny's Newsletter is Lenny Rachitsky's publication and podcast on product management, growth and career questions for product people.
Explore Lenny's Newsletter