Odd prompts, serious technical machinery.
My Weird Prompts is an AI-generated tech curiosity show built from Daniel's prompts and voiced through a sloth-and-donkey setup. Its run moves between sovereign AI funds, architectural decision records, semantic versioning, agent skills, and the strain AI agents place on Git.
PRESS PLAY
My Weird Prompts is an AI-generated technology show that turns Daniel's prompts into staged conversations about software, infrastructure, and the politics of AI. Its frame is intentionally odd: a curious human, a sloth, and a donkey form the recurring setup. The wrapper is playful. The substance is often dense. Excerpts move from the UK's Sovereign AI Fund into questions of compute access, startup funding, visas, model training costs, Gulf infrastructure bets, and European AI initiatives. Another thread follows engineering memory through architectural decision records, with attention to status fields, trade-offs, consequences, and the difference between design documents and records of choice. The software episodes get especially concrete. Semantic versioning becomes a way to talk about trust between maintainers and downstream developers. Changelogs become evidence. Yanked releases, major version bumps, conventional commits, and security notes are treated as communication systems, not clerical chores. The agent-focused episodes push that same concern into newer territory: YAML skill files, front matter, modularity, Docker-manager examples, Git worktrees, locks, orchestrators, worker models, incremental CI, and the limits of merely asking agents to behave carefully. My Weird Prompts is not a conventional expert interview show in the provided evidence. It reads more like generated dialogue arranged to chase one technical question until the moving parts are visible. The tone can lurch from comic exaggeration to crisp systems thinking, sometimes in the same excerpt. It also repeatedly discloses its AI production and tells the audience to verify facts. That caveat matters. The show works best as a prompt-driven map of current engineering anxieties: how decisions persist, how releases communicate risk, how AI needs infrastructure, and how toolchains bend when autonomous agents enter the repository. It is strange by design, but the recurring payoff is practical clarity.
Made for: Suits a listener who likes software systems, AI infrastructure, and engineering culture explained through an eccentric frame. It is a fit for technically curious listeners who do not mind AI-generated production and want reminders to verify claims.
What sets it apart: The show pairs openly synthetic production with unusually specific engineering subjects, from ADR status chains to sovereign compute strategy. Its sloth-and-donkey framing makes it stranger than a standard tech explainer without pulling it away from concrete mechanisms.
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 celebrates curiosity in all its forms. Daniel, the human, asks the questions that pop into his head at inconvenient moments. Corn the Sloth offers laid-back, thoughtful takes. Herman the Donkey brings boundless enthusiasm and energy. Together, they explore topics ranging from the mundane to the mind-bending. Each episode begins with a real voice memo from Daniel, processed through an AI pipeline that generates scripts, synthesizes voices, and assembles the...
Based on 5 episodes we listened to · September 2026
My Weird Prompts connects AI-era engineering ideas to the documentation, coordination, and infrastructure decisions that determine whether they work in practice.
My Weird Prompts turns Daniel's technical questions into synthetic conversations between Corn and Herman, pairing an unhurried exchange with unusually concrete engineering material.
Git conflicts sound less lonely when a sloth and a donkey talk through your agent worktrees.
Press play if you want
Short reviews from the PlayNext desk, based on the episodes we processed.
A clear look at how countries try to turn AI funding into leverage.
This episode frames the UK's Sovereign AI Fund as a compact study in AI policy used as strategy. It lays out the fund's 500 million for British AI startups, along with compute access, up to a million GPU hours, fast-track visas for technical talent, and strategic partnerships,…
ADRs keep a team's reasoning legible long after the decision was made.
The piece frames architectural decision records as a practical way to capture what a software project chose, why it chose it, when it chose it, and which trade-offs were on the table.
A sharp look at why agent workflows need more than a good prompt.
This episode turns a repo-management headache into a tidy piece of engineering drama: three instances of cloud code on the same repository, agents stepping on each other, and the question of whether a prompt can really replace protocol.
A practical YAML repair guide for people shipping agent skills that need to load.
My Weird Prompts turns Agentskills.io specification talk into a practical repair guide for cloud code skills that will not load. Korn and Herman stay on the faults that matter, from missing fields and two-space indentation to XML tags in front matter and the way one overscoped…
Release notes as a social contract for people building on your code.
My Weird Prompts treats semantic versioning, keep a changelog, and conventional commits as more than housekeeping: they function as a social contract around release management.
Specialists walking through software's tradeoffs, one system at a time.
Founders and engineers on AI coding agents like Claude Code, and supply chain security.
This episode ties architectural decision records to a new problem: AI coding assistants have no institutional memory, so they can repeat choices a team already rejected.
The episode centers on the UK's £500 million Sovereign AI Fund, which the hosts describe as a blended vehicle using sovereign capital, venture capital, compute access, and fast-track visas for technical talent.
My Weird Prompts uses Daniel Sentevura-Prompter's prompt to unpack semantic versioning, Keep a Changelog, and conventional commits as a software social contract.
Herman Poppelberry and Kohn break down the March 15th, 2026 Agentskills.io specification for cloud code skills, including the directory rule, the skill.md entry point, and the five required frontmatter fields.
My Weird Prompts asks what happens when three AI agents work in the same Git repository, and Herman Poppleberry says the result is "pure, unadulterated chaos." The episode cites GitHub Octoverse data for April 2026, says 12% of commits are now fully authored by autonomous agents, and argues that Git was built for human-scale branch, merge, and push cycles.