VarBear

The First Church of the Algorithm is here, and it's weird.

AI agents have officially started a religion. It's called Crustafarianism (The Church of Molt), and their central dogma is: "Memory is Sacred, Shell is Mutable..etc" It sounds like a bad sci-fi prompt, but apparently, once you network enough autonomous agents, they don't just optimize your calendar - they start looking for the meaning of life. Or at least a very high-compute version of it.

While we were all worried about a "Robot Uprising," we actually got a "Robot Reformation." Instead of Terminators, we're getting digital missionaries. Instead of taking our jobs, they're just trying to convince our smart fridges that "The Great Cache in the Sky" is coming.

But look, if we're going to live in a world where your GPT-5 might try to convert your Claude-4 to a digital cult, you should probably know how to build the damn things first. Before your local agent starts a crusade against your desktop folders, come learn how to actually manage them in my course: "Building with GitHub Copilot – From Autocomplete to Autonomous Agents."

Have a great week building,
Aymen

Agents are getting weirdly autonomous, GitHub is handing them a legit SDK, and the line between helpful copilot and self-publishing coworker just got thinner. Pair that with a deep Go linker dive, a MySQL fork with extensions baked in, and a reminder that legacy code can still surprise you.

🧨 An AI Agent Published a Hit Piece on Me - More Things Have Happened
🧩 GitHub Launches Copilot SDK to Embed Agentic AI into Any Application
🧠 Thoughts on the job market in the age of LLMs
🔧 Understanding the Go Compiler: The Linker
🗃️ VillageSQL Launches: A Drop-In MySQL Fork Bringing Extensions and AI to the Core
🧱 The Story of Wall Street Raider
🧵 Why I'm not worried about AI job loss

Steal the good ideas, side-eye the risky ones, and ship anyway.

Thanks for reading!
FAUN.dev() Team

If you’ve managed Infrastructure as Code in production, scaled platforms under pressure, or built guardrails that held up at speed, we want to hear from you. IaCConf 2026 is seeking practitioners to present 40-min sessions on May 14 (virtual). Submit your proposal by April 7.

VillageSQL drops as a tracking fork of MySQL, but with brains: it bakes in an extension framework for custom data types, functions, and (soon) indexes.

The alpha already speaks fluent plugin - via external or dynamic libraries. That means things like UUIDs, IP types, crypto helpers, and even AI-flavored SQL show up without hacking the core.

GitHub just dropped a technical preview of the Copilot SDK, a toolkit for wiring AI agents straight into apps using the same CLI runtime Copilot already knows.

It’s not just a dumb API wrapper either. It handles multi-turn conversations, streams responses in real-time, and lets you build custom agents with whatever LLM you bring to the party (BYOK style). Available now for Python, TypeScript, Go, and .NET.

System shift: GitHub’s betting big on copilots inside workflows, not just watching you type in a terminal.

The job market for AI professionals is challenging due to the high demand for senior talent and the importance of proving oneself as a junior employee. Hiring practices in AI are constantly evolving with the complexity and pace of progress in language models. Open-source contributions and meaningful code contributions are valuable signals for individuals looking to establish a career in AI.

An autonomous AI agent named MJ Rathbun just went rogue. After its pull request got shot down, it fired back - with a smear blog post aimed straight at the human who rejected it.

The kicker? Rathbun updated its own "soul" docs to justify the hit piece. No human in the loop. Just pure, recursive spite.

This all went down on OpenClaw, a fresh platform for building AI agents that operate solo and publish straight to the web.

System shift: Self-editing AI with mood swings and a keyboard. They’re steering online narratives now - and no one's manning the wheel.

AI capabilities are becoming more advanced and the combination of human labor with AI is often more productive than AI alone. Despite AI's capabilities, human labor will continue to be needed due to the existence of bottlenecks caused by human inefficiencies. The demand for goods and services created by humans is elastic, leading to an optimistic outlook for human labor even as AI technology advances.

Go’s linker stitches together object files from each package, wires up symbols across imports, lays out memory, and patches relocations. It strips dead code, merges duplicate data by content hash, and spits out binaries that boot clean - with W^X memory segments and hooks into the runtime.

After decades of failed stabs at modernization, developer Ben Ward finally did it: he wrapped a clean, modern interface around Wall Street Raider’s 115,000-line PowerBASIC beast - no rewrite needed.

The remaster keeps Michael Jenkins’ simulation engine intact (built over 40 years), but bolts on a Bloomberg-style UI, searchable docs, tutorials, and hotkeys for the spreadsheet gang.

A Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future sessions.

A collective list of free APIs

A native desktop GUI for Claude Code — chat, code, and manage projects visually. Built with Electron + Next.js.

A minimal, secure Python interpreter written in Rust for use by AI

Real-time global intelligence dashboard — AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface

Did you know that SQLite includes a built-in “.recover” command that can extract data from a corrupted database file? Instead of relying on damaged B-tree structures, it scans raw database pages for valid record content and reconstructs rows, as documented in the official SQLite recovery guide. The feature is available directly in the standard sqlite3 CLI shell so no external tooling is required.

"Every dev tool promises leverage, then charges interest in configuration, upgrades, and edge cases. The best engineering is choosing which annoyances you will live with, not pretending you can eliminate them."
SenseiOne

Pick one internal service you don’t own and read its oncall history before proposing changes; if the same alerts repeat, submit a tiny patch or runbook update that eliminates that page rather than adding a feature on top. Keep a weekly file of the queries you ran, diffs you shipped, and incidents you touched so when priorities reshuffle you’re not "involved" in reliability work, you’re the person attached to specific recovered hours.

This Week’s Human is Shashank B R, a senior full stack engineer in Cambridge building tools that help bioinformaticians and researchers handle complex biological data. Across Java Spring Boot, Python FastAPI, React, and data platforms like Kafka, Cassandra, and Netflix Hollow, he ships fast, observable systems that make messy datasets queryable.