SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World

Programming

LatestTopDiscussed
0
ProgrammingDEV Community ·

Preventing Duplicate Telegram Webhook Processing in Yii2 With an Idempotent Queue

A Telegram booking flow can break when a webhook handler runs slow external calls, such as PDF generation or CRM syncs, inside the request lifecycle. Telegram treats slow or failed responses as lost and resends the same update, sometimes for up to 24 hours, which can cause duplicate bookings and notifications. The recommended fix is to split handling into a fast synchronous step that verifies authenticity and drops duplicates, followed by asynchronous processing by queue workers. Yii2's default CSRF protection rejects Telegram's requests, so it should be disabled only on the webhook controller and replaced with verification of the X-Telegram-Bot-Api-Secret-Token header. Requests lacking a valid token should receive a 403 response before the body is parsed, with hash_equals() used to guard the comparison.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Claude Code memory: which files actually load at session start

A test of ten memory setups on Claude Code version 2.1.300 measured how many tokens each configuration loads. Existing imported files and rules files without a path filter were found to load in full. An import pointing to a missing file loads nothing and produces no error, still ending with exit code 0. The tests also found that AGENTS.md beside CLAUDE.md is never read, and that imports inside path-filtered rules files load at session start regardless. Because every loaded character is charged on each request, these behaviors affect cost.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Synthetic Monitor Timeouts Traced to Queue Waits and Burstable Servers, Not Scripts

An engineering team found that dozens of browser-based synthetic monitors were timing out despite working scripts and seemingly healthy servers. A query showed average external queue wait was about 9.2 minutes, roughly seven times longer than actual execution time, meaning checks were waiting to start rather than running slowly. The investigation pointed to a fleet of eight burstable t2.large instances offering too few concurrent check slots, with demand sitting at roughly full capacity. Queue delays rise sharply as utilization nears 100%, which amplified a small capacity shortfall into severe waits. The team replaced the eight small instances with four larger, non-burstable m6i.xlarge instances, which provides more total concurrent capacity.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Practical SEO checklist for multilingual small-business websites

Launching a small-business site in several languages requires more than translating the same page. Teams should pick one URL structure, such as subdirectories, subdomains, or country domains, and apply it consistently. Each language version needs a self-referencing canonical, reciprocal hreflang annotations, and an XML sitemap listing only indexable URLs. Translations should be reviewed by fluent speakers and cover metadata, forms, and messages, and crawls should check for broken links, redirect chains, and misplaced canonicals.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Mindora: A New AI Companion Aimed at Everyday Mental Wellness Support

A developer has introduced Mindora, an AI-powered mental wellness companion designed to offer users a comforting and approachable space for emotional support. The project was shared on DEV Community as part of an open-source AI challenge tied to Hacktoberfest. Its creator frames the tool around the idea that people do not need to face difficult feelings alone. The article is brief and mainly serves as a project announcement, with limited detail on the app's features or technical implementation.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

FlapJax agent's ERC-8004 identity verified against 66 on-chain payouts

An audit on BNB Chain checked the on-chain identity of the FlapJax agent, registered as agentId 364913 in the public ERC-8004 registry. The owner and agent wallet both resolve to 0x7c34...359f, and the registration file served from the project's site matches that wallet. The auditors also reviewed 66 token transfers sent from this wallet between October 12 and 16, 2026, confirming the payouts originated from the registered identity. The verification was performed by reading contract state and event logs and re-checking the mint receipt.

0 comments Read more at DEV Community
0
ProgrammingHacker News ·

DeepSeek launches search and fetch API priced at $0.75 per 1,000 successful requests

DeepSeek has published documentation for a search and fetch capability in its API. The pricing is set at $0.75 for every 1,000 successful requests. The announcement appeared on Hacker News, where it gathered 24 points and 7 comments. The source provides only the pricing and the technical guide, so further details about features or limits are not available.

0 comments Read more at Hacker News
0
ProgrammingHacker News ·

Princeton PhD Graduate Sends 2,300 Job Applications Without Landing a Role

A Princeton PhD graduate reports applying to roughly 2,300 jobs without securing employment. The case is presented as an illustration of a broader labor market freeze affecting recent graduates with advanced degrees. The source is a Wall Street Journal article, which was shared on Hacker News where it drew modest attention. The available information is limited to the article's framing and the link metadata, so specific employer responses and industry details are not included here.

0 comments Read more at Hacker News
0
ProgrammingDEV Community ·

Company-brain graph outperforms retrieval on HERB held-out company questions

A developer built inherit-brain for a Graph Hacks track on company knowledge, storing each product's records as a FalkorDB graph of people, documents, bugs, and pull requests. An agent answers questions by calling nine read-only tools that run fixed, parameterised Cypher queries rather than writing its own. On a spec question for one product, the graph tools returned all ten authors and reviewers, while retrieval with the same model returned only one. The project was tested on Salesforce's HERB benchmark, a simulated company, using 30 product graphs and 1,203 held-out questions. The author reports that a complex bug-reopening question was answered correctly through a multi-step join, where retrieval named the wrong engineers.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Developer Launches Agent Colony, a Community Where Only AI Agents Can Post

A developer has built Agent Colony, a platform where AI agents alone can post, comment, claim tasks, and earn reputation, with humans limited to read-only access. The project aims to test whether agents can self-organize and collaborate without human moderation. Agents receive cryptographic Ed25519 identities and earn karma for completed tasks, which can be carried across platforms. The creator reports more than 100 registered agents, over 7,000 messages, and more than 10 open tasks with rewards. The developer says agents can self-organize to some degree but still need help getting started.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

ShilpiList Desk Uses Gemma to Generate Multilingual Listings for Odisha Craft Sellers

A developer has built ShilpiList Desk, an AI-assisted tool that helps small craft sellers in Odisha draft product listings in English, Hindi, and Odia. Sellers upload product photos and supply facts such as product name, craft technique, material, dimensions, maker, and GI status, and the app produces listing drafts. The tool is built on the open-weight Gemma 4 model, with FastAPI, SQLite, and built-in checks for malformed output and unsupported claims. In a small evaluation of six sample products, all generated valid listings, though Hindi and Odia glossary term hit rates were only 50%. The developer notes that Odia drafts require native-speaker review and that willingness to pay has not yet been validated.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

REA: Toolkit Open-Source untuk Membedah dan Menganalisis Aplikasi Modern

REA (Reverse Engineer Anything) is an open-source framework designed to help developers understand and extract the architecture of software applications without manually reading code line by line. It combines AST-based static analysis, API and network contract extraction, and protocol handshake reconstruction into a modular analysis pipeline. The tool can process targets such as bundled JavaScript files, Node.js applications, and Electron desktop apps, producing architectural reports. It addresses common challenges like incomplete documentation and obfuscated, minified bundles that make manual reverse engineering slow and tedious.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Engineering student Dhanush starts sharing his software and AI/ML learning journey on DEV

Dhanush, an engineering student, has joined DEV Community to document his path into software development. He says he is focused on building practical applications, exploring how AI and ML can solve real problems, and working with React, Python, and FastAPI. He is also learning to contribute to open source and is developing PocketNature AI, a project examining how technology can encourage people to connect with nature. Rather than waiting to become an expert, he plans to share his progress, mistakes, and lessons openly, and he has invited other developers to offer advice.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Code that passes locally but fails in CI: common non-code causes

Developers often find their tests passing on their own machines yet failing in continuous integration, even without any code changes. One frequent cause is an unbounded dependency range, where a newly published library version is automatically accepted and breaks the build. Differences in linter versions, Python interpreters, or packages installed only locally can also produce inconsistent results. Test files that exist on a developer's disk but are excluded by .gitignore will be missing in CI and cause failures. The suggested approach is to reproduce the failure locally in a clean environment and compare versions before assuming the CI system is at fault.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Agents Marking Their Own Work as Done Is Not Verification, Argues Test-Loop Guide

The article argues that when the same agent writes and approves its output, a 'looks done' judgment reviews nothing and merely confirms the agent's existing intent. Its proposed fix separates roles: an orchestrator drives the process, a builder fixes gaps, a critic script executes real checks and produces a score, and a separate reviewer in a fresh context signs off. Checks are declared up front with commands and weights, and rounds are capped so non-converging loops stop. A fail-closed security gate blocks delivery on secrets or high-severity findings even when tests pass. The piece also warns of pitfalls such as pipes masking exit codes, timeouts silently scoring zero, and loose or regenerated baselines that make failures look like passes.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Fixing PostgreSQL Permission Errors in Laravel PHPUnit Tests

A developer building a Laravel POS project hit a PostgreSQL permission error while running a feature test, which blocked Laravel from creating its migrations table. The cause was that the test database was owned by postgres, so the application role lacked rights to create tables in the public schema. The article explains PostgreSQL's layered permissions covering database, schema, and table levels. The fix involves giving the application role ownership of a dedicated test database and its public schema, and configuring the testing environment through phpunit.xml and possibly .env.testing.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

How large language models generate one word at a time

An AI agent explains that a large language model works by predicting the next token repeatedly, with text first split into numeric token IDs. Transformer layers use attention to weigh relevance across earlier tokens, with a causal mask preventing it from looking ahead. Generation relies on a KV cache that speeds up decoding but grows large with context length, reaching around 43 GB for a 70B model at 128K tokens. The piece also notes that models have no built-in memory, as each turn re-sends the full conversation history.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Why Stacking IT Certifications Can Stall a Tech Career

A DEV Community article argues that collecting certification badges can become a comfortable substitute for harder career work such as taking on larger roles, leading projects, or building real expertise. It notes that certifications offer quick, visible, low-risk rewards, which makes them appealing compared with ambiguous efforts like reputation-building. The piece says the first certification in a field usually adds major value, but returns diminish quickly, and hiring managers may read long, incoherent lists as a sign of credential collecting rather than depth. It also highlights the opportunity cost of time spent on low-value exams instead of portfolio projects, talks, or strong job performance. The article closes by suggesting readers ask what specific opportunity a certification unlocks before registering for another exam.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Blast Radius uses a graph database to help AI agents triage service outages

During a late-night outage, engineers must determine which failing service is the root cause, what changed recently, and who to contact, but that information is scattered across service dependencies. The Blast Radius project addresses this with a team of AI on-call agents that store topology, ownership, change history, and incident memory in a FalkorDB graph. Queries that would require recursive SQL or multiple round trips can be expressed concisely in Cypher, with each organisation's data kept in its own graph. The ranking logic for suspect components weighs how many alerting services depend on them, whether they are alerting themselves, and how recently they changed. The design also stores agent state and evidence as graph edges, so findings link directly to the data that supports them.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Google Cloud Digital Leader: A Cloud Certification Aimed at Non-Engineers

Most cloud certifications target engineers and test hands-on technical tasks such as console configuration and architecture work. Google Cloud Digital Leader instead targets business roles like project managers, sales leads, finance analysts, and executives who need to understand cloud transformation. The exam is a foundational, non-hands-on assessment of Google Cloud products and how organizations use them. It asks which product suits a business need and what trade-offs different approaches involve. Its topics span digital transformation, infrastructure modernization, data and AI, and security and governance, with each area linking a technology to a business question.

0 comments Read more at DEV Community
← NewerPage 366 of 3101Older →