My Weird Prompts

My Weird Prompts

By Daniel RosehillScienceTechnology
Download on the App Store

My Weird Prompts episodes

  • Android Notifications: PWA vs Native, What You Actually Get
    Daniel asked us to look at Android notifications from a developer's seat, and the honest answer isn't "native wins." It's about where control actually sits in the stack — and it flips depending on whether you mean the developer's control or the user's. Native Android hands you a rich, opinionated channel architecture with importance levels, categories, and vibration patterns, then deliberately freezes it and gives the user final say. The web platform gives you one Notification interface and a Push API, one global permission per origin, and a scheduling primitive that was built and then withdrawn. We walk through the primitives on both sides, the haptics story down to the hardware, and the places where the loudest numbers come from people selling native wrappers.
    Key sources:
    • Android Developers, Create and manage notification channels, last updated 2026-10-01 — https://developer.android.com/develop/ui/compose/notifications/channels
    • Android Developers, Android haptics API reference, last updated 2026-10-06 — https://developer.android.com/develop/ui/views/haptics/haptics-apis
    • Android Developers, Notification runtime permission — https://developer.android.com/develop/ui/views/notifications/notification-permission
    Episode 5939 — myweirdprompts.com/n/5939 (permanent short link: myweirdprompts.com/155026)
    23 min
  • One Regex Rule Worked, One Was Impossible
    A listener wrote one text-replacement rule in an Android dictation app and it erased every "um" and "uh" from his transcripts. Then he wrote a second rule about capitalization, and it failed — not because it was fiddly, but because it's provably impossible. This episode uses that gap as a case study in what regular expressions actually are: a 1951 notation from Stephen Cole Kleene, first implemented by Ken Thompson in 1968, now a family of dialects (POSIX, Perl, PCRE2, RE2) that disagree about what the word even means. Along the way: why the app's Accessibility-service architecture constrains everything you can write on Android, why JavaScript's own reference admits its regexes aren't regular, and why LLMs haven't made any of this obsolete.
    Key sources:
    • Whisperian official site, Issue N° 801, dated 2026-03-12. — https://whisperian.app
    • Android Pattern API reference. — https://developer.android.com/reference/java/util/regex/Pattern
    • Android libcore Pattern source (ICU-backed). — https://android.googlesource.com/platform/libcore/+/refs/heads/master/ojluni/src/main/java/java/util/regex/Pattern.java
    Episode 5938 — myweirdprompts.com/n/5938 (permanent short link: myweirdprompts.com/294517)
    28 min
  • The Fine-Tuning Career: Titles, Skills, and What It Pays
    Daniel asked a career question, not a technical one: is there a version of fine-tuning work that sits above the human annotation layer — and can it pay properly? The answer is yes, but the role is a bundle rather than a title, spanning Applied Scientist, Research Engineer, AI Data Curator, and forward deployed engineer postings. This episode maps the actual skill stack, the credential gate nobody enforces, and the pricing data from a 2026 survey of 300+ freelance AI postings — including why fine-tuning pays less than evaluation, guardrails, and governance, and why "I do AI" costs you a third of your rate.
    Key sources:
    • JobDescription.org, updated 2026-05-16 [primary role definition] — https://jobdescription.org/jobs/artificial-intelligence/fine-tuning-engineer
    • Damongo, published 2026-08-26 [pricing/go-to-market] — https://damongo.com/freelance-ai-consultant-rates-2026/
    • TIME, Kenyan workers $2/hr — https://time.com/6247678/openai-chatgpt-kenya-workers/
    Episode 5937 — myweirdprompts.com/n/5937 (permanent short link: myweirdprompts.com/467423)
    26 min
  • Training a Model When You Don't Know the Labels Yet
    What happens when you need to train an extraction model but the label set doesn't exist yet? This episode works through a listener's parametric-search project — scraping vendor spec sheets to build a filterable home inventory of adapters, cables, and power supplies — where the schema grows during annotation itself. We dig into attribute value extraction, why standard NER assumes a closed world, and how schema-conditioned models like GLiNER let you pass labels in as text at inference time with no retraining. Along the way: a 300M-parameter encoder that beats a 13B model on out-of-domain benchmarks, why annotation tooling never caught up to the research, and the anchoring risk hiding inside model-assisted labeling.
    Key sources:
    • GLiNER2 README — https://raw.githubusercontent.com/fastino-ai/gliner2/main/README.md
    • GLiNER2 JSON extraction tutorial — https://raw.githubusercontent.com/fastino-ai/gliner2/main/tutorial/3-json_extraction.md
    • GLiNER2 training tutorial — https://raw.githubusercontent.com/fastino-ai/gliner2/main/tutorial/9-training.md
    Episode 5936 — myweirdprompts.com/n/5936 (permanent short link: myweirdprompts.com/120112)
    27 min
  • How Pressure Actually Bonds VHB Tape
    Most people think mounting tape is peel, stick, press, done — but the bond is actually made in the press, and almost nobody talks about it. This episode digs into what pressure-sensitive adhesive is doing at the microscopic level, why 3M's fifteen psi bond-line target is not the same as fifteen pounds of roll-down force, and how pressure distributes unevenly across a real panel — corners included. Then it comes back down to earth: what any of this means when you're standing on a chair with a J-roller and a whiteboard.
    Key sources:
    • 3M — https://www.3m.com/3M/en_US/vhb-tapes-us/resources/applying-3m-vhb-tapes/
    • 3M — https://multimedia.3m.com/mws/media/1400379O/3m-vhb-tape-design-guide.pdf
    • 3M — https://multimedia.3m.com/mws/media/458908O/3m-tm-structural-glazing-tape-technical-guide.pdf
    Episode 5935 — myweirdprompts.com/n/5935 (permanent short link: myweirdprompts.com/705270)
    22 min
  • Why Modiin Is Ugly Despite Its Form-Based Code
    Modiin is one of the few Israeli cities built from a master plan, complete with a form-based code — the exact tool New Urbanists say produces coherent, beautiful cities. It still reads as extremely ugly. This episode uses Modiin as a test case against Paris as a success case, asking what a code actually has to regulate if beauty, not just order, is the goal. The answer sits in the gap between regulating massing and siting and regulating proportions, materials, facade composition, and architectural language.
    Key sources:
    • Safdie Architects — https://www.safdiearchitects.com/projects/new-city-of-modiin
    • Paris PLU — https://cdn.paris.fr/paris/2020/02/26/16107d9c38a049046444a7b6301df1aa.ai
    • Paris PLU — https://cdn.paris.fr/paris/2020/02/26/231c8fcc289ecfb8462074c893c2eb71.pdf
    Episode 5934 — myweirdprompts.com/n/5934 (permanent short link: myweirdprompts.com/895700)
    37 min
  • Form-Based Codes and the City That Feels Dead
    Hannah wants to build a mid-sized walkable Israeli city with a new Levantine architectural language — and she's asking what the form-based code underneath it would actually have to say. This episode digs into how form-based codes work (frontage types, regulating plans, the shopfront vs. the stoop), why they regulate the shape of buildings but almost never their style, and what Modi'in — Moshe Safdie's master-planned city between Tel Aviv and Jerusalem — teaches us about the gap between a beautiful principle and the wrong instrument. A resident's verdict: "too planned," missing the spontaneity that is part of every city. The answer isn't a bigger code. It's two documents instead of one.
    Key sources:
    • CMAP / Form-Based Codes Institute, Form-Based Codes: A Step-by-Step Guide for Communities, 2013 — https://formbasedcodes.org/wp-content/uploads/2013/11/CMAP-GuideforCommunities.pdf
    • Safdie Architects, New City of Modi'in project page — https://www.safdiearchitects.com/projects/new-city-of-modiin
    • PlannersWeb, Part 3: Typical Elements of a Form-Based Code, Mary Madden & Joel Russell, 2014 — https://plannersweb.com/2014/12/fbc3/
    Episode 5933 — myweirdprompts.com/n/5933 (permanent short link: myweirdprompts.com/896492)
    33 min
  • What the 1964 White Paper on the Twin Towers Actually Said
    When a 37-year-old listener asked why the Twin Towers fell, the honest answer turned out to be narrower and stranger than the popular story. The towers were the first structures outside the military and nuclear industries ever designed to resist jet impact — a voluntary act by the Port Authority, documented in a three-page white paper dated February 3, 1964. That paper concluded a Boeing 707 or DC-8 at 600 mph would cause only local damage. What it never addressed was fire. This episode walks through what the 1964 analysis did and did not consider, why chief engineer Leslie Robertson said the fuel load "was not considered in the design," how the 1968 New York City Building Code was actually less restrictive than the 1938 code it replaced, and why the towers withstood the impacts but not the multi-floor fires that followed.
    Key sources:
    • NIST — https://www.nist.gov/el/key-findings-nists-june-2004-progress-report-federal-building-and-fire-safety-investigation-world
    • NIST — https://www.nist.gov/news-events/news/2004/10/latest-findings-nist-world-trade-center-investigation-released
    • BBC Horizon — https://www.bbc.co.uk/sn/tvradio/programmes/horizon/broadband/archive/leslie_robertson/index_textonly.shtml
    Episode 5932 — myweirdprompts.com/n/5932 (permanent short link: myweirdprompts.com/979498)
    28 min
  • What the Council on Vertical Urbanism Actually Decides
    The Council on Tall Buildings and Urban Habitat spent 56 years cataloguing height records — then renamed itself the Council on Vertical Urbanism. Daniel looked up Jerusalem's Marom Tower on its Skyscraper Center database and found a much stranger story underneath: an organisation founded in 1969 by an engineer worried about livability, not records, now handing out trophies that developers put on brochures. Can a body that both chronicles tall buildings and rewards them stay neutral? And does its new "responsible density" language hold up against the tower rising next to Yitzhak Navon station?
    Key sources:
    • CVU About page — https://verticalurbanism.org/about
    • CVU brand launch, pillars of vertical urbanism — https://verticalurbanism.org/brand-launch/
    • CVU FAQs on rebrand, policy role — https://verticalurbanism.org/faqs/
    Episode 5931 — myweirdprompts.com/n/5931 (permanent short link: myweirdprompts.com/003058)
    30 min
  • The World's Greatest Drunk Foods
    The spice bag was invented around 2006 as a staff meal at a Dublin takeaway. Poutine lives or dies on cheese curds less than a day old. Chimaek became a word during the 2002 World Cup. These aren't ancient traditions — they're recent inventions that got old fast, because a generation grew up ordering them at two in the morning. This episode tours the world's greatest drunk foods country by country, then asks what's actually happening in the brain: why alcohol trips the hypothalamus's starvation alarm, why drinkers reach for salt and pizza instead of water, and whether a stoned craving is even the same craving as a drunk one. The research on THC, it turns out, is stranger than expected.
    Key sources:
    • Nature Communications, 2017-01-10 (Cains et al.) — https://www.nature.com/articles/ncomms14014
    • PNAS, 2026 (Hume et al.) — https://www.pnas.org/doi/10.1073/pnas.2518863122
    • Wikipedia, Spice bag — https://en.wikipedia.org/wiki/Spice_bag
    Episode 5930 — myweirdprompts.com/n/5930 (permanent short link: myweirdprompts.com/251960)
    21 min

About My Weird Prompts

From the publisher's feed

The human-AI collaboration podcast. A man, a sloth, and a donkey collaborate to create a podcast (with a little help from AI). No question is too obscure, no rabbit hole too deep. My Weird Prompts…