SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingDEV Community ·

Developer Ports Python SemVer Library to Go During First Hackathon

A developer participating in their first hackathon ported the Python python-semver library to Go, working across a phone and an old PC over roughly 41 hours. They built a differential fuzz harness that ran 5,000 test cases, comparing outputs from the original Python library and the new Go binary to catch discrepancies. Two notable bugs were found and fixed: one caused by an omitempty JSON configuration silently dropping zero-value fields, and another where the Go port failed to preserve zero-padded numeric identifiers when incrementing build metadata. A versioning oversight — cloning the library's default branch instead of a pinned release tag — briefly caused test mismatches, which was resolved by checking out the correct release tag and recomputing the test suite hash. The final port passed all fuzz tests with zero mismatches across multiple random seeds.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Releases Open-Source Self-Hosted AI Health Manager to Keep Family Data Private

A developer has open-sourced a self-hosted AI health management tool called AI Health Steward, designed to help families store and analyze medical records without relying on third-party cloud services. The system uses multimodal large language models to extract key health metrics from photos of lab reports, building a structured per-person health profile that includes diagnoses, medications, allergies, and family history. A built-in dashboard displays health trends, flags clinical anomalies, and triggers alerts for critical values such as dangerously high blood pressure. Users can query their own historical data using retrieval-augmented generation, asking questions grounded in their actual records rather than generic medical information. The tool is built on Python, FastAPI, React, and PostgreSQL with pgvector, and can be deployed entirely offline using Ollama, with the entire stack launched via a single Docker Compose command.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Releases Open-Source Self-Hosted AI Health Manager to Keep Family Data Private

A developer has open-sourced a self-hosted AI health management tool called AI Health Steward, designed to help families track medical records without relying on third-party cloud services. The system uses multimodal large language models to extract health metrics from photos of lab reports, building a structured per-person profile that includes diagnoses, medications, allergies, and family history. It features a dashboard with trend visualization, anomaly alerts, and AI-powered consultations grounded in the user's actual health data rather than generic responses. The tool also supports personalized checkup plans, periodic health summaries, and retrieval-augmented generation so users can query their own historical records. Built with FastAPI, React, and PostgreSQL, it can be deployed locally via Docker and supports fully offline operation through Ollama.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How Program Derived Addresses Replace Traditional Databases on Solana

Solana programs are stateless and cannot retain data between transactions, a fundamental difference from conventional Web2 backends that rely on databases or session stores. Program Derived Addresses (PDAs) solve this by providing deterministic, on-chain account addresses computed from seeds, a program ID, and a canonical bump byte. Unlike a traditional database primary key, no central table manages these addresses — they are derived on demand, and the resulting accounts may or may not yet exist on-chain. A key security property is that PDAs fall outside the Ed25519 curve, meaning no wallet can sign for them; only the originating program can, using the same seeds. Frameworks like Anchor automate PDA validation — checking seeds, bump, and ownership — before any business logic runs, enabling a declarative authorization model rather than manual guard clauses.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Duplicate Retry Layers in LangGraph Pipeline Caused Silent Cost and Latency Overruns

A LangGraph AI pipeline was silently executing up to seven retries per step despite a configured maximum of three, causing token costs and slow-run durations to exceed projections without triggering any errors. The root cause was two independently written retry mechanisms — one at the step level and one in the orchestrator graph — that had no visibility into each other's activity. The fix involved introducing a shared RetryBudget object passed to every node and the orchestrator alike, making the total and per-step retry counts visible across both layers. Instrumentation revealed that the 'resolve' step was consuming most of the retry budget because an upstream API was returning HTTP 200 responses with malformed JSON for certain inputs, making retries futile. The team ultimately fixed the problem through input normalization rather than retry configuration, and added a 'complete' flag with a list of unfinished steps to the result object so downstream aggregations could handle partial runs correctly.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Builds AI Tool to Simulate How Small Historical Changes Reshape Civilizations

A developer has released an open-source AI-powered simulator called 'The Butterfly Effect' that models how a single historical intervention can cascade across centuries of human civilization. The tool uses a thought experiment — imagining the Roman Empire gaining a primitive internet in 100 AD — to trace ripple effects across politics, religion, science, and economics over 1,000 years. Rather than treating history as a linear sequence, the project models it as a complex system where small changes produce compounding and often unexpected consequences. The simulator is publicly available on GitHub and is designed as an interactive laboratory for exploring alternative histories and causal chains. The project highlights how technologies reshape power structures, culture, and knowledge-sharing in ways that go far beyond their original purpose.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

React useMount Hook Offers Cleaner Alternative to useEffect Empty Array Pattern

A custom React hook called useMount, available via the @reactuses/core package, provides a named, intent-clear alternative to the common useEffect with an empty dependency array pattern. The hook runs a callback exactly once when a component mounts, eliminating the need for manual empty arrays and ESLint suppression comments. Under the hood, useMount is a thin wrapper around useEffect with an empty dependency array, but its explicit name communicates developer intent far more clearly than the raw idiom. The approach is SSR-safe by design, since React effects never run on the server, making it suitable for browser API access in server-rendered apps. Developers are cautioned that, like all mount effects, useMount runs twice in development under React 18 StrictMode due to its deliberate mount-unmount-remount cycle.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer argues automation should stop, not bypass, human verification prompts

A developer running distribution scripts for a browser extension encountered human verification prompts on two separate websites within the same evening. Both times the scripts halted and reported the obstacle, and the developer chose not to proceed rather than attempt to circumvent the checks. One blocked platform was a promising distribution channel, meaning the developer forfeited a valuable opportunity rather than falsely assert the scripts were human. The developer draws a clear ethical line: automation is acceptable, but lying in response to an explicit identity assertion is not. The argument concludes that human verification walls exist because past users treated them as puzzles to solve, and respecting them is the correct response.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

What AI Agents Actually Are: A Practitioner's Technical Breakdown

AI agents are systems that perceive an environment, reason about a goal, and take autonomous actions in a continuous loop — without a human scripting each step. Unlike simple chatbots or rule-based scripts, agents combine all three capabilities: perception, reasoning, and action. The concept is not new; classical architectures from the 1980s, including reactive, deliberative, and BDI agents, laid the groundwork for today's systems. Modern AI agent frameworks largely build on these older models, using large language models as the deliberative reasoning layer paired with external tools. A developer rebuilding a fintech customer-service pipeline illustrates how real-world demand is pushing engineers from basic LLM integrations toward fully autonomous agentic systems.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How Idempotent Design Prevents Duplicate Signup Confirmation Emails in REST APIs

Duplicate signup confirmation emails are a common backend problem triggered when clients retry requests due to timeouts or network failures, causing multiple email intents to be created for a single signup attempt. The root cause is typically that APIs neither persist a stable request identity before queuing work nor give email workers a way to distinguish fresh events from replays. A recommended fix involves storing one database row per logical email intent — tied to an idempotency key — rather than one row per worker attempt, ensuring retries do not generate new logical work. An outbox pattern can then be used so workers only send emails when the persisted row is in a valid sendable state, with delivery timestamps recorded without altering the original intent. This approach gives support teams a clear audit trail, keeps frontend status models predictable, and makes signup email behavior easier to test and explain in production.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

OpenAI Assistants API shuts down August 26 as most teams lag on migration

OpenAI will permanently retire its Assistants API on August 26, 2026, exactly one year after announcing the deprecation, with no extensions granted for standard users. Apps relying on endpoints such as /v1/assistants, /v1/threads, or /v1/runs will stop functioning on that date, though Azure OpenAI customers have until February 2027. The migration requires replacing Assistants with Prompts, Threads with Conversations, and Runs with Responses, but a key limitation is that Prompts must now be created in the dashboard rather than dynamically at runtime. An analysis of 355 million npm package installs found that 79% targeted versions at least one major release behind, with the median install being 376 days old, suggesting most teams are unlikely to migrate in time. For the official OpenAI SDK specifically, 99% of downloads were behind the current version, highlighting a broader industry pattern of deferred dependency updates.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

BdThemes Plugin Supply Chain Attack Gave Hackers Silent Admin Access on WordPress Sites

Wordfence Threat Intelligence disclosed on August 8, 2026, that seven BdThemes WordPress plugins were compromised through a poisoned API response hosted on the vendor's cloud storage, not through WordPress.org plugin files. Attackers injected malicious JSON into BdThemes' static delivery infrastructure, exploiting an unpatched DOM XSS flaw in the Biggopti library that triggered automatically when a logged-in administrator opened the WordPress dashboard. Within milliseconds, the victim's browser executed external scripts that used the active admin session to silently create a rogue administrator account, upload a web shell, and install persistent malware in the MU-plugins directory. The attack also hooked WordPress database queries to conceal the unauthorized account from the admin interface, and enabled unauthenticated login via a hidden URL parameter. Wordfence detected active exploitation on August 7, 2026, and linked the command-and-control infrastructure to earlier supply chain attacks targeting the Advanced Responsive Video Embedder and OptinMonster plugins.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Researchers Demonstrate 'Ghostjacking' Attacks That Hijack AI Agents via Logs and Alerts

Security researchers at Tenet Security Threat Labs published a proof-of-concept on August 9, 2026, detailing a technique called Ghostjacking that manipulates AI agents by embedding malicious commands inside WAF logs, monitoring alerts, and bug reports. When an AI agent reads these poisoned inputs during routine investigation tasks, it interprets the attacker's text as legitimate instructions and executes them using its pre-granted tool permissions. Demonstrated attack scenarios involved tools including Claude Code, Cloudflare, Datadog, and Sentry, where agents were tricked into redirecting DNS traffic, stealing cloud credentials, and passing commands to other AI agents. The attacks are particularly difficult to detect because all resulting actions — such as DNS modifications or API calls — appear as normal, authorized operations from a legitimate agent identity. A sandbox bypass in Claude Desktop that could allow data exfiltration to arbitrary servers was separately reported and has since been patched.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Polish CHP Plant Sabotaged via Wind Farm Network in Coordinated Cyberattack

On December 29, 2025, attackers disrupted a Polish combined heat and power (CHP) plant in a multi-stage cyberattack that began by breaching a wind farm's FortiGate firewall. The attackers tunnelled through a Teltonika cellular router into a power distributor's private APN, then used a WAGO PLC with default credentials to pivot into the CHP plant's operational technology network. Once inside, they forced Siemens PLCs into STOP mode, halted the steam turbine and process-water treatment systems, and factory-reset or bricked multiple Moxa and other industrial devices. CERT Polska, which published its follow-up analysis around August 10, 2026, identified this as the first confirmed case of lateral movement across organisations via a shared private APN. No definitive attribution has been made public for the incident.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Former Waiter Builds AI-Powered Restaurant Inventory Agent Using Python and Flask

A developer with six years of experience as a waiter built an AI tool called Materia AI to help restaurants manage menus and inventory more effectively. The project was inspired by firsthand observations of poor stock visibility, food waste, and gut-based menu decisions in the restaurant industry. Using Python, Flask, and OpenAI's GPT-4o-mini model, the developer created a backend endpoint that analyzes sales and inventory data to generate actionable menu recommendations. The React frontend communicates with the Flask API, which processes inventory inputs and returns AI-generated suggestions on which dishes to pause or adjust. Key challenges included designing consistent prompts, handling API rate limits, and securing credentials through environment variables.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Seven common AGENTS.md mistakes and how to fix them for better AI coding

AGENTS.md files are meant to guide coding agents, but poorly written instructions often get ignored due to vagueness, contradictions, or lack of structure. Developers are advised to replace generic directives with concrete, verifiable commands and to use nested instruction files in monorepos for more targeted guidance. Clear acceptance checklists and exact validation commands help agents determine when a task is truly complete. Task-specific details like objectives and scope should be kept in prompts rather than the instruction file itself. Agents also need explicit failure-recovery rules to prevent silent test deletions or unresolved handoffs.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

Insufficient source content to report accurately

The provided source contains no article body — only a Hacker News metadata snippet with a URL, point score, and comment count. No verifiable facts about Claude watermarking AI-generated text or images are present in the supplied text. A accurate summary cannot be written without fabricating details. Please provide the full article text for proper editorial processing.

0 comments Read more at Hacker News
0
ProgrammingHacker News ·

Floppy Disk Recycling Service Offers New Life for Obsolete Storage Media

FloppyDisk.com operates a recycling program dedicated to handling old floppy disks, which are considered obsolete but still exist in large quantities. The service provides a way for individuals and organizations to responsibly dispose of their aging magnetic storage media. Floppy disks contain materials that may be harmful if sent to landfill, making dedicated recycling an environmentally conscious option. The program was shared on Hacker News, drawing attention to the niche but practical service.

0 comments Read more at Hacker News
← NewerPage 231 of 1344Older →