SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingHacker News ·

DeepMind Launches Dedicated Research Institute

DeepMind has established a new entity called the DeepMind Institute, signaling a formal expansion of its research operations. The institute appears to be a dedicated hub for advanced scientific and AI research under the DeepMind umbrella. The announcement gained traction on Hacker News, drawing community attention to the new initiative. Further details about the institute's specific focus areas and structure are available on its official website. The move reflects DeepMind's continued investment in organized, long-term research efforts.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

How Repository Guardrails Make Small AI Coding Models Surprisingly Effective

A software developer has shared how smaller, faster AI models like ChatGPT's Luna can handle real coding work when paired with strong repository-level constraints. Instead of relying on large prompt libraries, the developer uses structured documentation, naming conventions, and modular patterns within the codebase itself to guide the model. Tools like AGENTS.md, TypeScript's self-documenting features, and reusable skills help reduce the reasoning burden on lightweight models. The developer uses Codex inside VS Code to maintain visibility over code changes, especially on projects where code ownership matters. A key limitation noted is that smaller models struggle with ambiguity, requiring clearer task definitions compared to more capable frontier models.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

PS5 Linux Project Lead Resigns Over AI-Generated Code Quality Concerns

The lead developer behind a project to run Linux on the PlayStation 5 has stepped down from the role. The developer cited frustration with contributors submitting low-quality, AI-generated code produced by large language models they do not fully understand. This trend, described as an influx of inexperienced users relying on LLM tools, was given as the primary reason for the resignation. The departure highlights growing tensions within open-source communities over the impact of AI coding assistants on contribution quality and project maintainability.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

Google launches Gemini 3.8 Live and 3.5 Transcribe for real-time voice apps

Google has released two new AI models via the Gemini API and Google AI Studio to help developers build real-time, voice-first applications. Gemini 3.8 Live and its Extended Thinking variant enable conversational voice agents capable of background reasoning, asynchronous tool calls, and multilingual support across 97+ languages. The Extended Thinking model has ranked first on Artificial Analysis' Speech-to-Speech leaderboard for complex, multi-step tasks. Separately, Gemini 3.5 Transcribe — launched last month — offers dedicated speech-to-text conversion across 85+ languages with a word error rate as low as 2.6% in non-streaming mode. Both models are accessible through the Gemini API, with integration support from partners including LangChain, LiveKit, and Vercel.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

Deep Dive: How Digital Textures Are Structured and Rendered

A technical article titled 'Anatomy of a Texture' explores the inner workings of digital textures used in computer graphics. The piece breaks down how textures are stored, sampled, and applied in rendering pipelines. Published on a personal GitHub-hosted site, it has attracted attention on Hacker News with early community engagement. The article appears aimed at developers and graphics enthusiasts seeking a low-level understanding of texture mechanics.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

Why filter(Boolean) Removes More Than Just Null and Undefined in TypeScript

JavaScript's filter(Boolean) is a common shorthand to clean arrays, but it removes all falsy values — including 0, false, and empty strings — not just null and undefined. This distinction matters in TypeScript applications where values like retryCount: 0 or notificationsEnabled: false are valid data, not missing entries. Additionally, filter(Boolean) does not act as a TypeScript type guard, so the compiler cannot infer that null and undefined have been eliminated from the resulting array. A more precise approach is to use an explicit type guard such as value !== null && value !== undefined, which accurately narrows the type to exclude only nullish values. For repeated use across a codebase, a named utility like isNotNil from a library such as is-kit can express the same intent more readably and safely.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Huawei Launches Agentic AI Cloud Solutions in Nigeria at Lagos Summit

Huawei introduced its Agentic AI Cloud solutions in Nigeria at an AI and Cloud Summit held in Lagos, targeting developers building advanced, task-executing AI applications. Unlike traditional AI models that simply respond to prompts, agentic AI systems can plan tasks, call APIs, manage workflows, and coordinate multiple tools while operating under defined permissions. The launch signals how cloud providers are building infrastructure to support these more complex, action-oriented AI applications. Nigerian developers could apply agentic AI across sectors such as banking, healthcare, agriculture, logistics, and government services. Huawei has projected that autonomous AI agents could represent over 90% of global AI token traffic by 2035, underscoring the company's long-term strategic focus on agent-based computing.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Indonesia's 2024 Ransomware Attack on National Data Center: A Disaster Recovery Case Study

On June 20, 2024, Indonesia's Temporary National Data Center (PDNS) was struck by Brain Cipher ransomware, a variant of LockBit 3.0, crippling the infrastructure managed by the Ministry of Communication and Information Technology. The attack paralyzed over 230 government services and forced airports to revert to manual operations. Attackers demanded an $8 million ransom, and the incident exposed critical gaps in the government's data backup practices. The breach has since been examined as a major case study for building ransomware-resilient disaster recovery systems using cloud-based tools and security frameworks.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

Essay explores what a workless world might look like based on existing evidence

An essay published on Asterisk Magazine's Substack argues that we already have clues about what a world without work could resemble. The piece draws on existing real-world evidence rather than speculation to examine the potential social and economic implications of widespread joblessness. The article was shared on Hacker News, where it attracted 10 points. No comments had been posted on the Hacker News thread at the time of submission.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

NPS vs CSAT: How These Two Survey Metrics Serve Different Business Goals

Net Promoter Score (NPS) and Customer Satisfaction Score (CSAT) are two widely used survey metrics that measure fundamentally different things. NPS uses a 0–10 scale to gauge overall brand loyalty and likelihood to recommend, producing a score between -100 and +100. CSAT, typically rated on a 1–5 scale, measures satisfaction with a single interaction — such as a support chat or checkout — and is reported as a percentage of positive responses. While NPS is best suited for tracking long-term relationship health and early churn signals, CSAT delivers faster, narrower feedback on specific product or service touchpoints. Most teams benefit from running both metrics in parallel, as they complement rather than replace each other.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Von der Leyen Warns Frontier AI Could Enable Unprecedented Cyberattacks

European Commission President Ursula von der Leyen used her 2026 State of the Union address to highlight the cybersecurity dangers posed by advanced AI models, warning they could enable hacking at a scale previously unimaginable. She cautioned that such powerful models could fall into the hands of hostile actors, signaling that AI safety, model security, and supplier accountability will remain central to EU policy. Her remarks did not introduce new legal requirements but reinforced the EU's existing governance direction, including the AI Act framework. The EU is also expected to release a dedicated AI cybersecurity plan in July 2026, addressing both the threats and defensive opportunities that advanced AI presents. Von der Leyen additionally called for international cooperation on model evaluation and early warning systems, with partners including Canada and the UK.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Build a Free Client Portal Using Google Sheets and Sheetrocket in 30 Minutes

Agencies and freelancers often struggle with scattered email threads and disorganized project updates, making it hard for clients to track progress. A tutorial published on DEV Community outlines how to build a structured client portal using Google Sheets and a free tool called Sheetrocket, requiring no coding or database setup. The portal includes a dashboard, milestone tracker, invoice history, document links, and a messaging form, all accessible to clients via a single private link. Sheetrocket offers ready-made templates for various business types, including agency, freelancer, and coaching portals, which can be connected to a copied Google Sheet template. The entire setup is estimated to take around 30 minutes, with the business owner retaining full control over the data through their own Google account.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer replaces free hit counter API with Firebase after mobile blocking issues

A developer building a static web app called Daily Doodle attempted to add a simple visitor hit counter using a free third-party API, countapi.mileshilliard.com. The counter worked correctly on desktop browsers but consistently failed to update on mobile devices, with no visible error shown to users. Investigation revealed the likely cause was the API's domain being blocked at the network level by ad blockers or mobile carrier filters. Switching to Firebase Realtime Database resolved the issue, as Google's infrastructure domain is rarely blocked by such filters. The developer shared a step-by-step setup using Firebase's REST API with server-side atomic increments, requiring no SDK or authentication under the free Spark plan.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How One Dev Built a Cheat-Resistant Browser Game Leaderboard with Cloudflare D1

A developer built a browser game called 195-0 Game using Astro, Preact, and Cloudflare Workers, where players draft five people for cabinet roles and run a fictional campaign against 195 nations. A key engineering challenge was making the public leaderboard trustworthy without relying solely on client-reported scores. The solution uses signed draft tokens and server-side move replay: the server reconstructs the game from a seed and action log, then issues a signed result token before any leaderboard submission is accepted. Most content pages are prerendered, keeping the site fast, while the server handles draft issuance, score verification, and shareable PNG result card generation. The developer acknowledges the system validates legal move sequences but cannot confirm human participation, meaning a bot could still submit high scores through valid gameplay paths.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Cisco Patches Critical Email Gateway Flaw Enabling Root Access via Single Email

Cisco has confirmed active exploitation of CVE-2026-76461, a CVSS 9.8-rated SQL injection vulnerability in the AsyncOS email parsing logic of its Secure Email Gateway appliances. The flaw requires no authentication and can be triggered by a single crafted email, allowing remote attackers to execute commands as root on affected devices. Both physical and virtual appliances running AsyncOS 15.5 and earlier are affected, with fixed versions available across the 15.5, 16.0, and 16.5 release lines. Cisco disclosed the vulnerability on September 15, 2026, and CISA added it to its Known Exploited Vulnerabilities catalog, setting a federal patch deadline of September 17, 2026. No workaround exists, and administrators are advised to upgrade immediately while treating gateway logs as potentially compromised, cross-referencing firewall and network records for signs of intrusion.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Sentence-Window RAG Technique Improves Retrieval Accuracy Without Large Dependencies

Sentence-window retrieval is a RAG design pattern that separates the unit used for retrieval from the unit used for interpretation, indexing individual sentences while expanding results to include surrounding context. This approach addresses a core trade-off in standard RAG systems, where small chunks improve precision but strip away qualifying information, while large chunks restore context but introduce irrelevant material. A tutorial from Gate of AI demonstrates the technique using only Python's standard library, avoiding unverified third-party dependencies. Research cited in the tutorial found that multi-level chunk retrieval improved AI-evaluated answer correctness by over 25% compared to single-level retrieval in a Glycoscience-paper evaluation. The method is best suited for documents where facts and their qualifications appear in close proximity, and less effective when relevant evidence is scattered across multiple sections or documents.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

PostgreSQL pgvector Graph RAG Tutorial Unifies Vector Search and Graph Traversal

A technical tutorial from Gate of AI outlines a PostgreSQL-native approach to structure-aware Graph Retrieval-Augmented Generation (RAG), addressing limitations of conventional flat vector retrieval. The design stores text chunks with embeddings, a canonical entity graph, and community summaries within a single PostgreSQL database, using pgvector for semantic search and relational edge tables for graph traversal. This consolidation aims to eliminate the need for separate vector, graph, and document store systems, reducing infrastructure complexity and consistency overhead. The approach emphasizes quality gates to reject poor extractions before storage, and explicitly models time to distinguish current, superseded, or negated statements. The tutorial is intended as a design and evaluation guide for engineering teams considering structure-aware RAG rather than a framework- or provider-specific implementation guide.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Framework Proposes Auditable Human-Agent Collaboration for Support Ticket Triage

A new design framework called the Collaborative Human-Agent Protocol (CHAP) aims to make AI-assisted support ticket triage transparent and accountable. The framework treats agent recommendations as events within a collaborative workflow rather than final decisions, ensuring human judgments and overrides are recorded as structured operational data. It defines clear decision boundaries, separating low-risk agent suggestions from actions that require human authorization. The tutorial is implementation-neutral, meaning engineering teams can apply it across their own choice of platforms, model providers, and identity systems. Its core goal is to give organizations a reliable audit trail showing how work moved through the system and where human judgment was applied.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How to Build a Python Evaluation Gate for LLM Fine-Tuning Workflows

A new technical tutorial from the Gate of AI series outlines how to build a local, auditable evaluation gate before fine-tuning large language models. The workflow validates JSONL training datasets, fingerprints data, detects holdout overlap, and scores model predictions before approving a model for deployment. The guide emphasizes that completing a training run alone is not sufficient evidence for releasing a model, making a structured release gate an essential engineering control. It is designed to work with the Python standard library only, keeping it provider-agnostic until an organization selects an approved inference path. The tutorial notes that fine-tuning is best suited for stable, measurable tasks and warns against using it as a substitute for real-time data retrieval.

0 comments Read more at DEV Community
← NewerPage 1201 of 3225Older →