A new open-source tool called Engrim gives AI coding assistants persistent, project-level memory stored in a local SQLite database rather than tied to any single vendor's cloud. It lets developers switch between tools like Claude Code, Cursor, Windsurf and Google Antigravity mid-project while retaining architectural decisions and constraints, using a hybrid search combining SQLite FTS5 keyword search with lightweight vector embeddings. The creator reports it was validated over 105 sessions on a 50,000-line trading codebase with zero regressions and no loss of context across model switches.
github.com
· 2026-09-07
Vibe coding, the practice of using AI language models to write software with minimal manual coding, is drawing scrutiny after Georgia Tech researchers linked dozens of security vulnerabilities directly to AI-generated code in just a three-month sample. Despite these risks, a survey of over 1,100 professional programmers found that 72 percent use AI coding tools daily, with AI-generated or -assisted code making up 42 percent of their codebases and expected to exceed half by next year.
engadget.com
· 2026-09-06
A software engineering essay, part of a series called 'Conquering Entropy,' examines how AI-generated code changes the trust relationships within engineering organizations—from trusting the ticket author and reviewing engineer to trusting test suites, CI/CD pipelines, and even AI-driven site reliability tools. The author argues that as AI agents make code generation faster and cheaper, teams need explicit accountability: whoever ships a PR should be responsible for fixing it if it breaks in production.
kaeruct.github.io
· 2026-09-06
A programmer has released Bespoke, a tongue-in-cheek statically typed language that requires every command to begin with a polite greeting and end with an acknowledgement, using the .charming file extension. Dubbed 'The Queen's Code,' it treats the compiler, output console, host environment and shared objects as parties deserving formal correspondence rather than blunt instructions. The project pokes fun at the terse, imperative tone of conventional programming syntax, and its creator notes the implementation is far less developed than its etiquette rules.
blog.hofstede.it
· 2026-09-06
Early benchmark testing shows OpenAI's GPT-6 Astra model identifies about 4% more actionable bugs than GPT-5.6 Sol and 22% more than Opus 5 in code review evaluations. The advantage widens significantly on harder cross-file reviews, where Astra beats Sol by 20% and Opus 5 by 33%, suggesting stronger ability to trace how a change in one file affects code elsewhere.
coderabbit.ai
· 2026-09-05
Microsoft and hardware partners will sell high-end PCs later this year running Windows 11 Pro pre-configured under a program called Project Zenith, which tweaks default settings to reduce clutter and bundles tools like Visual Studio Code, GitHub Copilot, PowerToys and Windows Dev Skills. These machines require at least 64GB of unified memory and 250 GB/s of bandwidth to run local 30-billion-parameter AI models, with AMD Ryzen AI Halo devices launching first.
techspot.com
· 2026-09-04
Microsoft has branded its developer-optimized Windows configuration Project Zenith, targeting new devices with 64GB or more of unified memory. AMD introduced the first such device at IFA, a compact PC built on Ryzen AI Halo chips, with more machines using other silicon expected in coming months.
theverge.com
· 2026-09-04
A developer's blog post examines a debate sparked by a talk on debugging a nine-month software bug, questioning whether 'simplicity' as commonly understood in tools like Unix pipelines actually prevents such problems. The author compares a Unix shell pipeline and an equivalent Clojure program for word-frequency counting, showing that while both appear simple on the surface, modifying either to preserve original word order reveals hidden complexity and rigidity in the Unix approach.
jyn.dev
· 2026-09-03
HPE has released patches for CVE-2026-73749, a critical buffer overflow in ArubaOS-CX that lets unauthenticated attackers send malformed packets to a daemon to gain elevated code execution. The bulletin also lists 23 other vulnerabilities, several rated high severity, affecting management and web modules across multiple AOS-CX release branches. One vulnerable version has already reached end of maintenance but still received a fix due to the severity of the flaw.
bleepingcomputer.com
· 2026-09-03
Zeke Hausfather, a climate scientist who relies on AI for research, spent eight weeks tracking 1,138 prompts he sent to Claude Code to estimate the electricity consumed at the data centers processing them. His findings suggest that AI agents, which perform multi-step autonomous tasks, draw significantly more power than the simple chatbot queries tech companies typically cite as low-energy examples.
fastcompany.com
· 2026-09-03
Dallas has fitted 50 bulk-waste collection trucks with 100 AI-powered cameras that photograph properties along their routes and assign each one a 'blight score' from 1 to 4 based on detected issues like overgrown grass, litter and structural damage. Since testing began in April, the system has flagged possible violations at more than 21,000 properties, with some homeowners receiving code enforcement notices that could lead to fines. The cameras are run by a company called City Detect, which says it blurs faces and license plates but has no control over how the city enforces the data.
techspot.com
· 2026-09-03
The developer has released Muse Spark 1.3, an updated AI model available now through Muse Code and the Meta Model API. The update focuses on sustaining longer, multi-step tasks, following complex instructions more reliably, and collaborating more actively with users by asking clarifying questions and checking in before consequential actions. A more advanced reasoning mode is expected soon pending additional safety testing.
research.meta.ai
· 2026-09-02
GoKawiil