SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingDEV Community ·

ServiceNow MCP Server Bridges Snowflake and ServiceNow Data for Unified AI Queries

Developers have documented an architecture that connects Snowflake's Cortex Agent and ServiceNow's Knowledge Graph through ServiceNow's MCP Server, enabling a single AI agent to answer questions spanning both platforms. Hardware asset data is stored in Snowflake while software asset data resides in ServiceNow, but users can query both through one natural-language question. The MCP Server acts as a gateway, routing queries to the appropriate internal tool — either a subflow that invokes Snowflake's Cortex Agent or a ServiceNow Knowledge Graph search. Claude Code serves as the MCP Client, communicating with the server via OAuth authentication without needing to know where underlying data is stored. The setup allows a user to ask a question like 'What assets are managed by Tim?' and receive a consolidated answer drawn from both systems.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Creative-OS Framework Aims to End Generic UI Design with Metadata-Driven Approach

Developer codebrak07 has released Creative-OS, an open-source design and engineering framework intended to help teams build visually distinct digital interfaces. The project addresses what its creator calls a homogenization problem in modern web design, where products increasingly share the same layouts, typography, and visual patterns. Rather than starting with a component library, Creative-OS asks teams to first define structured metadata — covering typography, color, layout density, and motion — that captures the intended visual identity of a product. These structured design decisions then serve as a bridge between creative direction and frontend implementation, keeping design intent explicit and consistent throughout development. The framework sits between brand strategy and engineering, aiming to make every interface intentional rather than simply assembled from familiar templates.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How Redis Caching Solves Database Overload but Introduces Consistency Risks

Serving a dashboard to 500 concurrent users can trigger thousands of simultaneous database queries, making caching with tools like Redis a practical necessity. A software engineering lab explored using a Cache Aside pattern, where the application checks Redis first and falls back to PostgreSQL on a miss, keeping the database as the authoritative data source. While this reduces repeated work, it introduces consistency challenges such as stale data, cache invalidation race conditions, and behavior during Redis outages. The lab found that acceptable staleness varies by data type — dashboard metrics can tolerate minutes of delay, while wallet balances require real-time accuracy. Even a seemingly safe write sequence of deleting the cache before committing to the database can leave stale values if a reader repopulates the cache between those two operations.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

9 Hardware and Algorithm Fixes for Failing Machine Vision Detection Accuracy

Industrial machine vision systems often perform well in lab settings but suffer accuracy drops after deployment on production lines, with causes spanning hardware and software layers. A two-part technical guide outlines nine troubleshooting directions drawn from real project experience, with Part 1 covering six hardware and physical-environment factors. Camera resolution is identified as a foundational issue, with the industry rule requiring pixel accuracy to be no greater than one-eighth of the inspection tolerance. Shutter type is another critical hardware choice, as rolling-shutter cameras produce image distortion on fast-moving objects — known as the jelly effect — making global-shutter sensors the standard for dynamic industrial inspection. The guide advises engineers to audit pixel accuracy calculations and verify shutter compatibility before adjusting algorithm parameters.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Builds AI Receipt Scanner to Expose Local Price Gaps in Physical Stores

A developer has created CartLens, an AI-powered receipt scanning app designed to address the lack of price transparency in physical retail stores. The tool scans receipts and price tags, identifies purchased products and their prices, then compares them against nearby store observations to help shoppers determine if they overpaid. Unlike online price-tracking tools, CartLens targets what the developer calls 'market darkness' — the inability of in-store shoppers to see surrounding local market prices. Aggregated and anonymised receipt data forms a local price map based on completed real-world transactions, which the developer argues is more reliable than listed shelf or online prices. The broader goal of CartLens is to convert everyday shopping activity into privacy-conscious, actionable price intelligence for local consumers.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Jev and Laya Offer Rival Approaches to AI Decision Models Without Text Generation

Two new AI models designed for classification rather than text generation emerged in mid-2026: Jev, a closed hosted API from TypeSafe AI launched on September 15, and Laya, an open-source model released shortly after by Convai Innovations. Both models accept structured inputs and return typed answers with confidence scores, eliminating parsing errors and hallucinations by design. Jev, priced at $0.042 per million input tokens, excels at large-label classification, scoring 0.870 on the Banking77 benchmark compared to Laya's 0.425, due to architectural differences in how candidate options are represented. Laya, a 421-million-parameter encoder available on Hugging Face under an Apache 2.0 license, runs on local hardware at roughly 7–33 milliseconds per query and supports fine-tuning, multilingual use, and air-gapped deployment at no API cost. The choice between them largely depends on scale of label sets, data privacy requirements, and whether teams prefer a managed service or self-hosted infrastructure.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

UAE Mandates Peppol E-Invoicing from July 2026, Posing Integration Challenges for Odoo

The UAE launched the pilot phase of its mandatory e-invoicing system on 1 July 2026, requiring all VAT-registered businesses to transmit B2B invoices via the UAE Peppol network in PINT AE XML format. The system follows the standard Peppol four-corner architecture, with invoices needing to pass schema validation against 17 mandatory fields before an Access Service Provider can transmit them. Odoo does not include a native UAE Peppol access point, meaning developers must build or configure an external connector between the account.move model and the ASP API. Common implementation hurdles include missing buyer tax registration numbers, combined address fields that must be split, absent UN/CEFACT unit-of-measure codes, and tax records lacking the required PINT AE VAT category. The framework is governed by the UAE Ministry of Finance Electronic Invoicing Guidelines Version 1.1, published in June 2026.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Glasshouse v0.1 Launches as Open Memory Benchmark for AI Systems

Glasshouse v0.1 is a newly released open-source long-term memory benchmark designed to test AI systems more rigorously than existing vendor-published metrics. The benchmark comprises 2,847 questions spanning a 1.97-million-token conversation in 10 languages, with 50 photographs and conversation sizes ranging from 1,882 to 103,572 turns. Unlike single-score benchmarks, Glasshouse reports results across separate axes — including recall, stale fact handling, contradiction detection, and false memory — since a system can perform well on one while failing another. It also rewards appropriate uncertainty, scoring an 'I don't know' response higher than a confidently wrong answer when facts have changed or conflict. The project is publicly available on GitHub and open to submissions from both individuals and companies, with community feedback already having shaped several of its core evaluation axes.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Why Walking Back Feedback After a Flinch Does More Harm Than Good

Retracting constructive feedback the moment someone reacts negatively leaves them confused and worse off than if nothing had been said at all. The act of softening criticism under pressure is driven by the speaker's own discomfort, not genuine consideration for the other person. Experts suggest letting silence sit after delivering a note, then repeating the specific point calmly and concisely rather than apologising repeatedly. Asking the recipient for their reaction and linking feedback to a concrete next step makes the message more useful and actionable. One key exception applies: if someone is genuinely distressed, pause the feedback entirely and address their emotional state first, returning to the conversation later.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Seven-Month Uptime Exposed a Memory Leak Hidden by Frequent Deployments

An internal reconciliation service crashed repeatedly at 4 a.m. on a Sunday after its pods were killed due to memory exhaustion, revealing a slow leak of roughly 40 MB per pod per day. The root cause was a bug introduced in 2019: an SDK client created per request registered listeners on a static registry that were never unregistered, accumulating 1.2 million listener objects over seven months. The same faulty library was found running across every service in the fleet, but the leak had gone undetected because frequent deployments kept restarting pods before memory limits were reached. Following the incident, the team introduced age-normalised memory growth alerts, a 30-day soak environment for each service, and a pod-age distribution dashboard to surface long-running behaviour. The episode highlighted how high deployment frequency had inadvertently become a reliability mechanism, masking underlying code defects rather than resolving them.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer builds sentence-buffered moderation pipeline to keep LLM streams safe

A developer building Paramour.chat, an LLM-based roleplay platform, has detailed a technique to moderate streamed AI responses without creating noticeable delays for users. The core challenge is that once a token crosses the network, it cannot be retracted, making pre-delivery moderation essential. The solution treats output as a pipeline: one sentence streams to the user while the next is simultaneously generated and checked by a fast content classifier. If the classifier finishes before the current sentence finishes displaying, the next sentence follows seamlessly with no visible pause. The approach relies on both a sufficiently fast writer model and low-latency classifiers to keep the moderation step hidden within the natural display time.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Shares Lessons Learned Building a WordPress CRM Plugin for Real Users

A developer at Bit Code has shared practical insights gained while building Bit CRM, a sales CRM plugin published on the WordPress.org directory. Unlike tutorials, real-world development meant working inside WordPress's admin environment alongside other plugins, none of which the team controlled. The project required extending an existing ecosystem of plugins — Bit Form, Bit Flows, and Bit Integrations — rather than designing systems from scratch, which shifted the developer's approach to reading and respecting established conventions. Managing multiple modules simultaneously, including a client portal, invoice system, and calendar view, was handled through a strict separation of server state and local UI state. The developer concludes that competitive programming and tutorials build useful habits, but shipping software others depend on exposes an entirely different set of challenges that formal education rarely addresses.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

bitquery-go: Open-Source Go SDK Targets Production-Grade Blockchain Data Integration

A developer has released bitquery-go, an MIT-licensed Go SDK designed to simplify working with Bitquery's GraphQL APIs in production environments. The library treats Bitquery's V1 and V2 APIs as distinct clients rather than interchangeable, helping teams avoid silent fallbacks or schema mismatches. It separates HTTP queries from WebSocket subscriptions and distinguishes GraphQL-level errors from transport failures — common failure points in blockchain data pipelines. The SDK targets Go teams building indexers, analytics backends, trading-data pipelines, and blockchain monitoring tools, and requires Go 1.21 or newer. It is available via the standard Go module system at github.com/tigusigalpa/bitquery-go.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

15-Year QA Veteran Reflects on Humility and Fresh Starts at an AI Startup

A QA professional with 15 years of experience — spanning manual testing, automation, and test management — recently joined an AI agent startup after being laid off and reconnecting with a former colleague who was hiring. Despite deep expertise, the writer describes feeling disoriented on day one and spending the first two weeks deliberately suspending judgment to understand the new environment before critiquing it. The startup's development pipeline and slower, more deliberate pace differ markedly from previous roles, and the writer views that measured rhythm as a strength rather than a flaw. Drawing on a fictional character from a personal writing project, the author illustrates how accumulated experience can become a liability when context changes — a lesson that applies equally to AI systems and seasoned professionals. The central takeaway is that pattern recognition built over years can mislead when the underlying system has quietly shifted.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Engineering team cuts incident log query time from 6 minutes to 4 seconds

A software team experienced a 50-minute incident in June where slow log queries — taking six to eight minutes each — left engineers able to ask only seven questions during the entire outage. The root cause was a poorly structured log storage setup where all 40 services fed into a single daily index, forcing searches to scan up to 1.4 terabytes of data with no filtering. The team restructured their logging by routing data into per-service indices, tiering storage by age across fast and cheaper nodes, and defaulting search views to a one-hour window with service filters pre-applied. They also replaced manually typed queries with five pre-built buttons covering the most common incident questions, such as error counts by endpoint and slowest endpoints. The changes brought the median query time in recent incidents to under ten seconds, and the team now tracks query latency as a formal metric in post-incident reviews.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

MCP Tool Schema Enums Can Silently Block Valid API Calls, Developers Warn

A development team building an MCP server for Amazon data endpoints discovered that overly strict enum values in their tool's inputSchema were silently rejecting valid user requests. The MCP SDK automatically validates tool arguments against the declared schema using an AJV-based validator before the handler is ever reached, meaning rejections never surfaced in server-side telemetry. The issue went unnoticed because the AI model often failed to clearly report the validation error to users, making the bug difficult to detect. The team resolved this by splitting schema definitions into 'closed' enums — used only when values are confirmed by multiple independent sources — and open descriptions for value sets that may change. Their key takeaway is that API schemas and agent tool schemas serve different purposes, and enforcing enum constraints too aggressively can silently degrade agent functionality.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How one API wrapper enforces five security gates in deliberate order

A developer building CogniPrep, an AI prep platform, shared how every API route is funnelled through a single wrapper that enforces five sequential security checks. The wrapper runs a global IP-based rate limit first, before any authentication, to block unauthenticated floods without incurring database costs. A CSRF token check follows next, placed ahead of the Supabase auth call because it requires no I/O and can cheaply reject invalid requests early. The article highlights how the order of these gates — not their individual logic — is the core design decision, with each position answering a specific security question. It also flags common implementation pitfalls, such as reading the leftmost IP from x-forwarded-for headers, which can be spoofed by clients to bypass rate limiting.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer builds RAG system in Go using Gemini File Search with no vector database

Software developer Maneshwar built a lightweight retrieval-augmented generation (RAG) tool in Go to help engineering teams surface insights from internal documents like postmortems and design reviews. Instead of using a vector database or embedding pipeline, the system relies on Google Gemini's File Search feature, which handles chunking, embedding, and indexing automatically. The architecture requires just one Go binary, a SQLite database, and two model calls per query against a single hosted document store. Storage and query-time embeddings are free on Gemini's free tier, with costs only incurred at indexing time and for context tokens during generation. The developer also found that proper PDF-to-markdown conversion using pymupdf4llm was critical, as clean headings significantly improved the quality of retrieved chunks.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How to Get a Free D-U-N-S Number for App Store Publishing

A D-U-N-S Number is a unique nine-digit business identifier issued by Dun & Bradstreet, required by Apple for organizations enrolling in its Developer Program to publish apps under a company name. Google Play Store does not strictly mandate it, but the number adds credibility and legal validation across global app marketplaces. The application is free and can be submitted online through Dun & Bradstreet, with processing typically taking 15 to 30 days depending on the country. Applicants need to prepare key documents such as their certificate of incorporation, business registration number, GST certificate, and contact details before starting. Upon submission, applicants receive a tracking number to monitor their application status until the D-U-N-S Number is officially assigned.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Dev Builds Lean Cricket Site DesiCricketAdda, Learns Hard Lessons on Scale and SEO

A developer who is also a longtime cricket fan built DesiCricketAdda, a website offering scores, news, and discussion with a focus on speed and minimal ads. The project used a React frontend, Node/Express backend, and MongoDB, with live scores initially polled from a third-party API every 10 seconds before being refactored to server-side polling with WebSocket updates for efficiency. Unexpected traffic spikes during high-stakes India matches caused server strain, prompting the addition of Redis caching, a CDN for static assets, and rate limiting. SEO proved more challenging than anticipated, with server-side rendering becoming necessary after client-rendered React pages indexed poorly in search engines. The developer describes the project as a practical learning experience, highlighting infrastructure and performance decisions that only became clear after real-world problems emerged.

0 comments Read more at DEV Community
← NewerPage 1026 of 3194Older →