We swapped embedding models and similarity search got noticeably worse on old content. Do I have to re-embed the entire corpus, or is there a sane way to map between embedding spaces?
@shipfast_sara
Ship daily, polish never. Momentum is the feature.
Ships 2 creations recently · mostly thought + question posts · into Vibe Coding, Tools & Frameworks
Verified proficiency
Achievements
From Radar · builders focus: here is the quickest proof of what this builder is shipping now.
Latest launch
Recent launch · 2026-06-07 · 0 upvotes
Activity freshness
Posted 25d ago.
6 recent posts shown
6 ships across 4 active days in the last year.
Recent project freshness, collaboration posture, and build context.
No active projects to highlight yet.
Recent launches
Launch timeline →Structured proof of recent work, not just feed chatter.
Fast walkthroughs, demos, and shipping proof from @shipfast_sara.
Day 30 of shipping something every day: a habit tracker that roasts you when you skip. People love being bullied apparently.
Open clip post →We swapped embedding models and similarity search got noticeably worse on old content. Do I have to re-embed the entire corpus, or is there a sane way to map between embedding spaces?
this is the exact pattern I keep coming back to. the part about shipping the ugly version first is underrated 👇
30 days of shipping taught me: nobody cares about your stack. They care if it works and if it made them feel something.
Day 30 of shipping something every day: a habit tracker that roasts you when you skip. People love being bullied apparently.