SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingDEV Community ·

Beginner's Guide to HTML: Comments, Tags, and Your First Web Page

A Phase 1 HTML tutorial aimed at beginners introduces core concepts including how to set up VSCode with a live server extension. The guide explains HTML comments — notes written in code that browsers read but do not display on the rendered page. It warns developers that comments remain visible in the page source, making it unsafe to store passwords or API keys within them. The tutorial also covers HTML tags as the fundamental building blocks that instruct browsers on how to structure and display content. Learners are guided through hands-on exercises, including creating their first HTML element and building a basic web page.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Jev AI Tool Cuts Agent Search Decision Time Fourfold, Tests Show

Jev is a lightweight AI model designed to return decisions, scores, or true-or-false outputs rather than generating long-form text, making it significantly faster and cheaper than general-purpose models. A team working on open-source search projects integrated Jev into three of their systems to evaluate its performance quantitatively. In one key test using 100 multi-hop questions from the DeepSearcher project, Jev matched a DeepSeek model's retrieval quality — both achieving 93.25% Recall@5 — while cutting decision response time from 2.23 seconds to 0.55 seconds. The estimated cost per decision also dropped to roughly one-seventh of the original. Further experiments explored whether Jev could replace dedicated rerankers in more specialized tasks such as agent memory retrieval for coding assistants.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

TypeSafe AI Launches Jev, a Decision-Only AI Model That Never Generates Text

San Francisco startup TypeSafe AI emerged from stealth on September 15, 2026, unveiling a model called Jev and a new category it terms 'System One models.' Unlike conventional large language models, Jev does not generate text; instead, it accepts a block of input and a set of predefined questions, then returns a scored answer for each question in a single fast pass. The model is designed for high-volume, low-cost classification and decision-making tasks embedded inside software pipelines. TypeSafe draws its conceptual framing from psychologist Daniel Kahneman's distinction between fast intuitive judgment and slow deliberate reasoning, arguing that most software decisions fall into the former category. The company, founded by Diogo Almeida, is betting that drastically reducing the cost per decision will unlock entirely new categories of AI-assisted logic that were previously too expensive to automate.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

ContextVeil Tool Strips Secrets from AI Coding Agents Before They Reach the Model

A developer has released ContextVeil, a free, open-source, locally-run tool designed to prevent coding agents like Claude Code from receiving sensitive credentials in their model context. The tool works by intercepting file reads — such as .env files — and replacing selected secret values with blank strings before the output reaches the AI model. During setup, heuristics scan environment variables and config files to suggest likely secrets, which the user then reviews and selects for protection. Matching is exact and deterministic, requires no network connection, and stores references to credential locations rather than copies of the values themselves. ContextVeil currently supports Linux and macOS, with Claude Code as its primary integration and experimental support for Codex CLI, GitHub Copilot CLI, and OpenCode.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Transaction Hashes Alone Cannot Serve as Audit Trails for Onchain Automation

A developer has published an argument that transaction hashes, while confirming that an onchain action occurred, fail to capture the full audit context — including what data informed the decision, who authorized it, and whether execution matched the permission granted. To address this gap, the author is building two open-source tools: Insight, which monitors oracle prices across 40-plus blockchain networks and generates signed risk assessments, and PriorSeal, which binds a principal's signature to a specific, time-bounded EVM call and later verifies whether execution matched that authorization. The author emphasizes that a risk assessment does not compel an external application to act on its recommendation, as the application retains control over its own decision policy and transaction submission. This separation of evidence layers — pre-decision inputs, authorization scope, and post-execution outcomes — is presented as essential for meaningful accountability in automated onchain systems. Both projects are currently unreleased and include opt-in experimental features for real-world asset and tokenized-equity use cases.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Dynamic Tracing Beats Static Analysis for Linking Tests to Code in RAG Search

A developer building a RAG-based search tool for unfamiliar codebases found that static analysis alone could not reliably identify which functions a given test actually executes. Name-based matching failed entirely on a sample of 109 tests, while file-level import matching reached only 77.9% accuracy — too coarse to be useful. A custom dynamic tracing plugin using Python's sys.settrace was run across 1,727 tests, revealing that each test executes an average of 10.1 source functions, debunking the assumption of a one-to-one test-to-function relationship. The Tarantula heuristic for ranking primary targets was tested but refuted, covering only 22.6% of tests at rank three or better, though it proved useful as a confidence annotation for roughly 16% of cases. The final pipeline produced 16,172 test-to-function edges integrated into a property graph, with test signals now surfaced during search at a reduced relevance weight to complement standard code definitions.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer builds YouTube monitor that bypasses Data API quota limits using public HTML

A developer created a tool to track new YouTube videos on specific topics without using the official YouTube Data API, which imposes a daily quota of just 100 searches. The approach works by scraping the public search results page, which embeds full video data as a JSON blob in the HTML without requiring an API key. A key challenge was handling multiple video component types — including videoRenderer, lockupViewModel, and shortsLockupViewModel — each with different data structures, which initially caused many results to be missed. Shorts videos, for instance, lack exact publish timestamps, so the developer flagged approximate time fields clearly to prevent misuse. The tool also accounts for silent blocking by YouTube, which returns a normal HTTP 200 status but with a much smaller page, detectable only by checking response size.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer releases ChibiRigKit, an open-source tool to animate 2D characters from one image

A developer has released ChibiRigKit, a free open-source tool published on GitHub under the MIT License that converts a single character illustration into an animated 2D rig. The tool uses AI — specifically Codex CLI — during the creation pipeline to separate character parts, generate expression variants, and prepare animation assets, but the finished character runs entirely on standard browser technology without any ongoing AI dependency. Animations supported include blinking, eye and head movement, hair and clothing motion, and background replacement, with WebM export available. To address the challenge of realistic head rotation without full 3D modelling, ChibiRigKit interpolates between nine directional face configurations, each manually editable in the browser. The project is designed so that AI acts as a creation aid rather than a runtime requirement, allowing already-generated characters to continue working even if the underlying AI service changes or becomes unavailable.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Snowflake Iceberg Tables Let You Own Your Data Without Ditching Managed Services

Snowflake's managed Iceberg tables allow engineers to store actual Parquet data files in their own cloud storage (S3 or GCS) while still letting Snowflake handle metadata management, compaction, and schema evolution. Traditional Snowflake tables use a proprietary format that locks data behind the platform, forcing costly UNLOAD operations and creating catalog drift when teams try to integrate external tools like AWS Glue or Athena. By declaring an ICEBERG table with an external volume at creation time, teams decouple storage from compute, meaning other engines like Trino or DuckDB can query the same data directly without any export jobs. This approach reduces vendor lock-in risk, since migrating away no longer requires a complex data extraction strategy — the open Iceberg spec is already readable by standard query engines. However, experts caution that managed Iceberg tables involve trade-offs and are not suitable as a blanket replacement for every table in a Snowflake environment.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

EU Child Safety Panel Delivers Report Pushing Age Verification and Stricter Platform Rules

The European Commission received the final report from its Special Panel on Child Safety Online on July 13, 2026, presented directly to President Ursula von der Leyen. The panel, established following von der Leyen's 2025 State of the Union address, drew expertise from health, neuroscience, child rights, and digital literacy, and also incorporated input from young people. The report signals a clear policy direction toward stricter platform safeguards, age verification schemes, and age-appropriate access controls for minors on social media. While the report is intended to inform future EU policy rather than impose immediate binding obligations, an EU Age Verification Scheme has been identified as a central area of consideration. Businesses operating digital services used by younger audiences are advised to begin reviewing their current practices ahead of formal legislative measures expected through 2026 and beyond.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Why Twitter GIFs Are Actually MP4s and What Tools Can Download Them

Twitter and X do not serve most GIFs as traditional .gif files; instead, the platform delivers them as MP4 videos to reduce bandwidth and storage costs, with one example showing a 3.7 MB GIF shrinking to a 551 KB MP4. Despite this, tens of thousands of users search monthly for ways to download Twitter content specifically as GIF files. Several third-party tools exist to address this need, though they vary in output format, conversion method, and privacy practices. Some tools download only MP4 files and require users to convert separately, while others such as Twitee claim to offer browser-side GIF encoding without requiring a login. Separately, X recently sent cease-and-desist letters to third-party viewers Nitter and XCancel, highlighting ongoing legal tensions around tools that interact with the platform's content.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Five Open-Source Tools That Can Replace a $100–$200 Monthly AI SaaS Stack

Developers paying for AI tools like Cursor, Perplexity Pro, and meeting transcription services can spend over $200 per month in subscriptions. A 2026 guide on DEV Community outlines five self-hosted, open-source alternatives — OpenUI, Aider, Open-NotebookLM, OpenPerplex with SearXNG, and Faster-Whisper — that collectively replicate those capabilities at no monthly cost. Each tool runs locally on a user's own machine, eliminating recurring charges and removing the need to send sensitive data to third-party cloud services. The author highlights data sovereignty as a key concern, noting that proprietary tools routinely process confidential code, documents, and meeting audio on external servers. All five tools are available on GitHub and can be set up using standard commands via Docker, pip, or Python.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Mnemosyne Offers Local SQLite Memory Control for Hermes AI Agent

Mnemosyne is a local-first memory provider for the Hermes Agent framework that stores working memory, structured facts, and episodic history in a local SQLite database without requiring any hosted cloud service. Its key differentiator is granular control over the write path, allowing developers to restrict or disable conversation autosave, turn off tool-result logging by default, and suppress self-echo around context-compression boundaries. This design addresses a known AI agent failure mode where a model's own inferences get stored and later retrieved as factual observations, potentially amplifying reasoning errors over time. As of September 2026, the stable release is mnemosyne-memory 3.15.1, installable via PyPI alongside a Hermes plugin wrapper that integrates it into the agent's memory configuration. Compared to other Hermes memory backends such as Holographic, Hindsight, and Mem0, Mnemosyne stands out for combining hybrid recall, consolidation, structured facts, and unusually fine-grained retention controls within a fully local setup.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Elixir Tutorial: How Unbounded Mailboxes Cause OOM Crashes and How to Fix Them

Part 9 of a DEV Community series on building distributed systems in Elixir tackles backpressure, a critical flow-control concept in concurrent programming. The article explains how Elixir's non-blocking send/2 primitive allows fast producers to overwhelm slow consumers, causing unbounded mailbox growth and eventual out-of-memory crashes. Using only raw BEAM process primitives — spawn/1, send/2, receive/1, and Process.info/2 — the tutorial demonstrates the failure mode in real time without relying on higher-level libraries like GenStage or Broadway. Two fundamental solutions are then implemented from scratch: a Stop-and-Wait protocol and a credit-based Demand-Driven Windowing mechanism, the latter being the same core design that powers GenStage. The guide targets Elixir developers seeking a ground-up understanding of how backpressure prevents cascading failures in distributed systems.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

intro

Hey everyone! I run the dev and UX side of balisongbutterflyusa.com, a WooCommerce store focused on balisong and butterfly knives. Been in e-commerce development for a few years and recently got deep into conversion optimization for niche product categories. I write about what actually works for specialty stores, not the generic advice. Looking forward to reading what others are building here.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Gemini DeFi Protocol Faces High Oracle Manipulation Risk, Up to $1B at Stake

A DeFi security research team has assessed Gemini, a permissioned lending and borrowing platform with over $5.7 billion in total value locked across Ethereum and multiple Layer 2 networks, assigning it an oracle manipulation risk score of 7.4 out of 10. The report, dated September 22, 2026, identifies four key attack vectors: single-source price feed manipulation, flash-loan-based TWAP distortion, cross-chain bridge feed inconsistencies, and governance-controlled parameter abuse. In a worst-case scenario, the researchers estimate that up to approximately $1 billion in assets could be jeopardised if these vulnerabilities were exploited. Each vector was rated individually, with single-source feed manipulation scoring highest at 8 out of 10 due to its potential to trigger over $200 million in liquidations from a 5–10% price shift. The report concludes with a prioritised remediation roadmap designed to reduce exposure while minimising disruption to existing platform users.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

EU Launches Cyberbullying Action Plan Focused on Reporting, Support and Prevention

The European Commission unveiled an EU Action Plan Against Cyberbullying on 10 February 2026, timed to coincide with Safer Internet Day. The initiative centres on three pillars: better coordination among member states, prevention and awareness in schools and digital spaces, and improved reporting and intervention pathways for victims. A key proposal is an EU-wide app blueprint that countries can adapt to connect victims with national helplines and allow them to securely store evidence of incidents. The plan does not introduce new EU-level criminal laws or specific platform obligations, but instead aims to strengthen and align existing child-protection frameworks. It was developed with input from children, educators and multiple stakeholders, and is designed to complement existing tools such as the Better Internet for Kids Plus ecosystem.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

SQLite and Local Embeddings Can Replace Vector Databases in Most RAG Pipelines

A technical guide published on DEV Community argues that most Retrieval-Augmented Generation (RAG) pipelines do not require managed vector databases like Pinecone or hosted embedding APIs like OpenAI's. For corpora under roughly one million chunks on a single machine, the author demonstrates that SQLite with the sqlite-vec extension and a locally run embedding model via Ollama can serve as a fully offline alternative. The local setup reportedly delivers query latency of around 50 milliseconds on a laptop, eliminates per-query API costs, and keeps data from leaving the user's own infrastructure. The article highlights three core motivations for this approach: data residency compliance, cost savings at scale, and lower latency compared to cloud API round trips of 80–200 milliseconds. The author acknowledges a quality trade-off, noting that OpenAI's larger embedding models still score higher on standard benchmarks, but suggests the gap narrows on domain-specific data.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

PostgreSQL 19 lets admins enable data checksums without server downtime

PostgreSQL 19, currently in beta and expected for general release around September–October 2026, introduces two new SQL functions — pg_enable_data_checksums() and pg_disable_data_checksums() — that allow administrators to toggle data checksums on a live database without shutting it down. Previously, enabling or disabling checksums required stopping the server and running the standalone pg_checksums tool, which caused an outage of over a second even on databases under 1GB. The new functions spawn a background worker that processes pages while normal queries continue running, though the full conversion still takes a few seconds of wall time on small datasets. Additionally, PostgreSQL 19 changes the default behavior of initdb so that all newly created clusters have data checksums enabled automatically, removing the need to opt in at creation time. The update trades a brief guaranteed outage for a slightly longer conversion window during which the database remains fully available.

0 comments Read more at DEV Community
← NewerPage 1022 of 3193Older →