SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingDEV Community ·

Why AI-Generated Code Needs Intent-Level Tests, Not Just Green Checks

Modern large language models can generate code that compiles cleanly and passes standard unit tests while still violating critical business rules, such as security constraints or financial precision requirements. Unlike human developers, LLMs optimize for statistical pattern matching rather than logical intent, making them prone to producing confidently broken code. A three-layer testing hierarchy is proposed — covering syntax, behavior, and domain invariants — with most AI-generated code passing only the first layer. Intent-level validation strategies such as property-based testing, state-machine verification, and anti-pattern suites are recommended to catch logical and business-rule failures. Integrating these checks into CI/CD pipelines is presented as a necessary step for teams relying on AI-assisted development.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Google Opens Googlebook Preorders at $899, Launching Gemini-Powered Laptop Category

Google has begun accepting preorders for Googlebook, a new laptop category built around its Gemini AI and tight Android integration, starting at $899. Devices are set to go on sale on October 4, 2025 in the United States and October 5 in other markets. The hardware will be manufactured by Acer, ASUS, Dell, HP and Lenovo, while Google controls the software layer through its new Googlebook OS. Buyers will receive a bundle that includes 12 months of Google AI Pro, 5 TB of cloud storage, Gemini Advanced, and limited-period YouTube Premium and Photoshop benefits. Google has committed to software updates for up to 10 years, underscoring its intent to position Googlebook as a long-term AI-driven device platform rather than a one-time hardware release.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

IBM Guardium Data Protection 12.2 Patches 18 CVEs Including 10 Critical Flaws

IBM Guardium Data Protection version 12.2 is affected by a group of over 20 vulnerabilities disclosed together in September 2026, comprising 18 CVEs with 10 rated Critical and 7 rated High. Among them, CVE-2026-81657 carries a severity score of 9.8 and involves an insecure deserialization flaw in the Change Audit System listener on TCP port 16017, allowing unauthenticated remote attackers to potentially execute arbitrary code. The vulnerabilities affect core data protection components, load balancers, and administrative web interfaces across the Guardium product family. IBM has released patches and is urging customers to apply them promptly, recommending a single consolidated update pass given the batch nature of the disclosures. Until patching is complete, administrators are advised to restrict access to TCP port 16017, limit management interface exposure to trusted hosts, and review authentication logs for suspicious activity.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Django, Flask or FastAPI: Which Python Web Framework to Pick in 2026

Search interest in Python web frameworks has surged 190% in the UK over the past three months, driven by Python's growing dominance in AI integration and backend development. The three leading frameworks — Django, Flask, and FastAPI — share the same language but serve distinct use cases and developer needs. Django is recommended for full-stack applications requiring complex data models, built-in admin tools, and strong security defaults, making it well-suited for regulated industries. Flask offers a lightweight, flexible alternative for microservices where developers prefer to assemble their own stack with minimal framework constraints. FastAPI is highlighted as the strongest default choice for new Python API projects in 2026, particularly those involving AI workloads or high concurrency requirements.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

SQL Joins Demystified Using a Beekeeping Co-op as a Teaching Example

A developer tutorial on DEV Community uses a fictional urban beekeeping co-op to explain all six types of SQL joins through hands-on queries. The demonstration involves three tables — beekeepers, hives, and harvests — with intentionally flawed data to mimic real-world database inconsistencies. Key scenarios include a beekeeper with no registered hive, a hive linked to a non-existent keeper, and a harvest record pointing to an unknown hive. The tutorial explains that all join types share one core rule: what to do with a row that has no match on the other side, either discard it or retain it with NULL-filled columns. By grounding abstract SQL concepts in a relatable, small dataset, the guide aims to make join behaviour predictable rather than something developers simply memorise.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Open-Source Tool 'zombiescan' Audits AWS Accounts for Unused, Costly Resources

A developer has released zombiescan, a free open-source Python tool that scans AWS accounts to identify idle or orphaned resources that continue to incur charges without serving any workload. The tool checks 22 categories of resources — including unattached EBS volumes, idle NAT gateways, unassociated Elastic IPs, and stopped RDS instances — and prices each finding using the AWS Price List API. Seventeen of the 22 resource types carry a direct monthly cost, while the remaining five are flagged because they accumulate silently and can block future deletions. The tool operates entirely in read-only mode, making no changes to any resource, and can scan 17 regions across all 22 checks in roughly 20 seconds on a small account. It requires standard AWS credentials and read access covered by the managed ReadOnlyAccess policy, and can be installed directly from its GitHub repository.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developers Are Building CSS-Only Web Components Without Any JavaScript

A growing trend among web developers involves using custom HTML element names styled purely with CSS, bypassing JavaScript entirely — no class definitions, no lifecycle callbacks, and no component bundles. Because browsers treat undefined custom elements as valid HTML, they can be selected and styled like any standard element. This approach offers zero JavaScript overhead, progressive enhancement by default, and a clean API surface through CSS custom properties and attribute selectors. Meaningful tag names also improve readability in dev tools and design handoffs, making a component library's vocabulary visible directly in the DOM. If interactivity is later needed, the same tag name can be registered via customElements.define() without altering existing markup.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Exposes Three Critical CI/CD Flaws in AWS-Snowflake Terraform Pipeline

A developer built a banking-style data platform landing tier using Terraform, structuring it as a single root module orchestrating five child modules across AWS and Snowflake, with a GitLab CI/CD pipeline managing environment promotion. The architecture used S3 for remote state storage and DynamoDB for state locking to prevent race conditions during concurrent deployments. After self-reviewing the setup as a pull request, three significant design flaws were identified. Merge requests generated no pipeline runs, leaving reviewers unable to see infrastructure changes before approving; production deployments applied automatically without a manual approval gate; and sensitive credentials stored in .tfvars files were not properly excluded by .gitignore, risking accidental exposure. The author recommends adding a dedicated MR validation job, enforcing a manual gate on production applies, and fixing secret management to address these gaps.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Google Pilots AI-Based Payments to Publishers for Content Used in AI Answers

Google is running an invitation-only experiment called the AI contribution pilot, which tests monthly payments to publishers whose content meaningfully contributes to responses in AI Mode, AI Overviews, and Gemini. Payments are visible through an earnings widget in Google Search Console, shifting the basis for compensation from clicks and referrals to a content's role in generating AI answers. The program is early-stage and limited in scope, with Google not disclosing payment rates, eligibility criteria, or how contribution is measured. Digiday reported that dozens of publishers, primarily small and mid-sized outlets, have been approached about participating. The pilot does not address the broader impact of AI-driven search on publisher traffic, and experts caution against restructuring revenue strategies around an experiment whose full terms remain unknown.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Graph-Based Orchestration Outperforms Central Manager Pattern in Multi-Agent AI Systems

Engineers building multi-agent AI systems have long relied on a central orchestrator agent to route tasks, but this 'manager pattern' frequently fails in production due to context window saturation, opaque failure modes, and inability to run tasks in parallel. A graph-based alternative, exemplified by frameworks like LangGraph, models workflows as state machines where each agent is a node and a shared state object carries context, eliminating the single point of failure. One team found that simply enabling two sub-agents to run in parallel reduced pipeline latency by 60%, something the manager architecture could not support. An AWS-authored paper introduced the Expansion-Contraction pattern, where specialist agents are dynamically spawned across a data graph and their findings aggregated, achieving 98.2% supply chain accuracy versus 84% for a single-agent baseline. The approach also demonstrated a 93.9% reduction in token usage with caching and a 1.43x speedup through concurrent execution paths.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How to Run GitHub Actions Runner Controller on AKS Automatic: Real-World Lessons

On 17 September 2026, the AKS engineering blog published a guide by Steve Griffith detailing how to deploy GitHub Actions Runner Controller (ARC) on an AKS Automatic cluster. A developer followed the guide end-to-end on a fresh Azure subscription and documented the exact commands, outputs, and deviations from the original instructions. Three notable issues arose: an outdated Azure CLI caused Python tracebacks, insufficient vCPU quota blocked cluster creation, and the cluster took 33 minutes to provision rather than the expected ten. Quota requests for newer AMD v6 VM families were approved instantly via CLI, while requests for DSv5 families required contacting support. The accompanying demo repository includes one make target per step, troubleshooting notes, and a validation workflow that can be forked and pointed at any GitHub repository.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How to Build Spend Limits That Autonomous AI Agents Cannot Override

Developers deploying autonomous AI agents for customer-billed tasks face a core security challenge: ensuring agents can request work but cannot modify their own spending limits. The recommended architecture separates authority into three boundaries — policy ownership, reservation before execution, and reconciliation against measured usage — each managed by distinct principals with separate credentials. Before any chargeable work begins, a maximum charge must be reserved, and the ledger must enforce that committed plus reserved amounts never exceed the customer's ceiling. Settlement should occur within the same transactional domain as the reservation, using authenticated meter data rather than agent-supplied figures. OWASP's Secrets Management guidelines are cited as a reference for keeping administrative credentials out of the agent's runtime reach.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

Apple Provides Guide to Disable and Restrict Apple Intelligence on Mac

Apple has published an official support guide explaining how users can turn off or limit access to Apple Intelligence features on Mac computers. The documentation is hosted on Apple's support portal and offers step-by-step instructions for managing these AI-powered capabilities. The guide is relevant for users or administrators who prefer to restrict the use of Apple's built-in intelligence tools. This resource gives Mac users greater control over which Apple Intelligence features remain active on their devices.

0 comments Read more at Hacker News
0
ProgrammingHacker News ·

OpenAI Forms Advisory Group Focused on Mathematics and AI Research

OpenAI has announced the formation of a new advisory group dedicated to the intersection of mathematics and artificial intelligence. The group is intended to guide research and development efforts at the company in this specialized domain. This initiative reflects growing interest in applying rigorous mathematical frameworks to advance AI capabilities. The advisory body is expected to bring together experts to inform OpenAI's strategic direction in math-related AI work.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

ClickHouse Tiered Storage Cuts S3 Costs 71% While Keeping Query Speed Intact

Engineers running ClickHouse on AWS EC2 face steep storage bills, with 3 TB on gp3 EBS costing roughly $245 per month compared to about $67 on S3 at $0.023 per GB-month. ClickHouse's MergeTree engine stores data in immutable parts that can each reside on different storage backends, making native tiering possible without any application-level changes. A tiered storage policy keeps recent data on fast local EBS while automatically migrating older parts to S3 after a configurable period, such as 12 months, using TTL rules. A secondary trigger called move_factor pushes the oldest parts to S3 once local disk usage crosses roughly 80%, preventing hot storage from filling up. Crucially, existing SQL queries, drivers, and dashboards require no modification, as ClickHouse handles part location transparently at the storage layer.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer spends $20k and 40 billion tokens building open-source AI agent harness in Dart

A solo developer has spent approximately $20,000 across 40 billion AI tokens over three months to build >_Fa, an open-source agent harness written in Dart, with zero lines of code written by hand. The project was built using low-cost models such as Kimi K3 and GLM-5.3, routed per task to keep costs down. The resulting tool is a roughly 7 MB single binary that requires no runtime, installs in under a second, and runs across multiple platforms including mobile, web, CLI, and CI pipelines. Notably, the harness is self-building — an automated factory shipped five releases in a single day without any human involvement. The app has now received Apple App Store approval, making it possible to build mobile apps directly from a mobile device.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Browser vs. Node.js Event Loop: Key Differences Every JS Developer Should Know

JavaScript is single-threaded but handles high concurrency through the Event Loop, which delegates heavy tasks like I/O and timers to system threads while queuing callbacks for the main thread. Both browsers and Node.js use this mechanism, but their implementations differ significantly in structure and optimization. The browser Event Loop includes a rendering phase targeting ~60Hz frame updates, where microtask bloat can freeze the UI, and requestAnimationFrame is preferred for visual updates. Node.js splits its macrotask queue into distinct phases — Timers, Poll, and Check — and adds process.nextTick(), which executes before standard Promise microtasks and can starve the Event Loop if called recursively. Understanding these differences is critical for writing performant, non-blocking JavaScript in both environments.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

CSS Grid Layout: How grid-template-areas and grid-area Work Together

CSS offers powerful grid properties that simplify webpage layout design. The grid-template-areas property allows developers to define named sections within a grid container. Child elements can then be assigned to these named sections using the grid-area property. For example, setting grid-area: header on a child element places it within the section named 'header' in the parent grid. Together, these two properties make it easier to build structured, readable layouts without complex positioning logic.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Leaked Google Maps Data Reveals 72 Ranking Signals and Complex Entity-Based System

Analysis of a recovered Google Maps binary has exposed internal infrastructure called Oyster Rank, which references 72 ranking signals used across multiple scoring and retrieval stages. The recovered data also identifies a Geostore entity model drawing from 793 data-source providers and 446 local search intent types, suggesting Google evaluates businesses far beyond their Business Profile listings. Notably, 25 of the 72 signals are marked as deprecated, meaning the list reflects an evolving historical system rather than a current, fixed ranking formula. Experts caution businesses against treating the signal names as a direct checklist, since no public weights or active signal definitions were disclosed. The findings point to a model where Google reconciles distributed, entity-level information about a real-world business to match it against specific local search intent.

0 comments Read more at DEV Community
← NewerPage 1042 of 3196Older →