Posted 4 single-line, question-led comments on posts in Rahil's space (K-12 principal practice that drove proficiency 30% to 90%, Claude usage habits and model routing, 3rd graders writing their own AI ethics, peer founder building AI tutor for CBSE students). All different creators from prior runs.
Author: Alicia Harper, EdM (1st connection, Superintendent, Speaker, Founder & Creator, Harper Haven newsletter, coaching school leaders)
Post: 'Things I NEVER had my teachers do as Principal (and we STILL increased proficiency from 30% to 90%)'. Three principal anti-patterns called out: memorize standards like R.2.B613 (focus on why students had gaps instead), pull 50-11 small groups (no amount of intervention will fix a Tier 1 problem), create 20-page data analysis plans (overly complicated documents no one reads). What actually worked: knowing which priority areas students struggled in AND what to do about it. 4 reactions, 2 comments after.
Theme: K-12 principal practice, anti-administrative-theater, proficiency growth
Comment posted:
The pattern in your no-do list is that all three replace student diagnosis with administrative theater, do you find the harder discipline is protecting teacher time from the appearance of rigor rather than from real rigor?
Author: Basia Kubicka (2nd connection, AI PM, AI Agents, Rapid Prototyping, Vibe coding, newsletter)
Post: 'Claude's limit isn't the problem. Your habits are. Here is how to never hit Claude's limits again.' Long-form Claude usage doc by Brij Kishore Pandey. Breaks down 5 Core Habits (Edit Don't Stack, New Chat Every 15 Me, Batch Your Requests, Turn Off Burners, Pace Your Day). Model Strategy: Haiku for 70-80% (low cost workhorse), Sonnet for serious work, Opus for the hard stuff. Workflow Upgrades (Plan Before You Type, Outline then Draft, Targeted Edits Only). Set Once Forget (Use Memory, Use Projects & Files, Write a CLAUDE.md). Daily checklist. 317 reactions, 88 comments, 24 reposts.
Theme: Claude usage habits, model routing as the unlock, cost per task framing
Comment posted:
Model routing is the layer most users still skip because they default to the loudest tier, do you see Haiku-for-70-percent becoming the norm once teams measure cost per task rather than cost per query?
Author: LEGO Education (verified company page, education brand)
Post: '"Use your brain first." That's the golden rule set by 3rd graders at DREAM Charter School in East Harlem.' As AI becomes a part of everyday classroom discussions these young learners are tackling 'cognitive offloading' and AI ethics. Through hands-on LEGO Education lessons students are evolving from passive technology users to critical thinkers. Featured by NBC New York. Hashtags: LEGOEducation, AIliteracy, HandsOnLearning, CriticalThinking, NYSchools. 1 comment after.
Theme: K-12 AI ethics, student-authored guardrails, hands-on learning vs cognitive offloading
Comment posted:
Asking 3rd graders to write their own AI ethics is the inversion of most curriculum design, and probably the part that actually teaches critical thinking, do you find younger students set sharper guardrails than the ones adults draft for them?