<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Pickuma</title><description>Editorial reviews of AI and developer tools. Tested, not generated. AI-assisted, human-supervised.</description><link>https://pickuma.com/</link><language>en-us</language><item><title>Cursor SDK Review: Building AI Agents With Known Limitations</title><link>https://pickuma.com/posts/cursor-sdk-review-building-ai-agents-limitations/</link><guid isPermaLink="true">https://pickuma.com/posts/cursor-sdk-review-building-ai-agents-limitations/</guid><description>Cursor&apos;s new SDK exposes the same agent runtime that powers the editor. We break down what ships, where the documentation lags, and when the limitations matter for production code.</description><pubDate>Tue, 12 May 2026 09:05:52 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>OpenAI Codex Chrome Extension: Browser-Native AI Coding Agent Tested</title><link>https://pickuma.com/posts/openai-codex-chrome-extension-browser-ai-agent/</link><guid isPermaLink="true">https://pickuma.com/posts/openai-codex-chrome-extension-browser-ai-agent/</guid><description>OpenAI&apos;s Codex Chrome extension puts its coding agent inside your browser tab. We tested the workflow patterns that pay off, the limits worth knowing, and how it fits next to Codex CLI and IDE agents.</description><pubDate>Tue, 12 May 2026 09:04:34 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>OpenCode vs Claude Code: Why 157K Developers Are Hedging Against Anthropic</title><link>https://pickuma.com/posts/opencode-vs-claude-code-157k-developers-hedge-anthropic/</link><guid isPermaLink="true">https://pickuma.com/posts/opencode-vs-claude-code-157k-developers-hedge-anthropic/</guid><description>A measured comparison of OpenCode and Claude Code, the lock-in math behind the split, and a decision framework for picking one, the other, or both.</description><pubDate>Tue, 12 May 2026 09:03:15 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Qwen 3.6 Plus API: Pricing, Benchmarks &amp; Developer Access Guide (2026)</title><link>https://pickuma.com/posts/qwen-3-6-plus-api-developer-guide-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/qwen-3-6-plus-api-developer-guide-2026/</guid><description>A measured developer review of Alibaba&apos;s Qwen 3.6 Plus API — pricing vs GPT and Claude, 1M-token context behavior, coding benchmarks, and the access paths that actually work.</description><pubDate>Tue, 12 May 2026 09:01:19 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>OpenAI Codex vs Claude Code: Hands-On Python Benchmark for Devs</title><link>https://pickuma.com/posts/openai-codex-vs-claude-code-python-benchmark/</link><guid isPermaLink="true">https://pickuma.com/posts/openai-codex-vs-claude-code-python-benchmark/</guid><description>We pointed Codex and Claude Code at the same Python codebase across refactoring, debugging, and agentic tasks. Here is what each tool shipped, where each one wins, and what the speed-vs-cost tradeoff actually looks like in practice.</description><pubDate>Tue, 12 May 2026 08:59:31 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>ModelScope Review: Alibaba&apos;s Model-as-a-Service Platform for AI Developers</title><link>https://pickuma.com/posts/modelscope-review-alibaba-model-as-a-service-platform/</link><guid isPermaLink="true">https://pickuma.com/posts/modelscope-review-alibaba-model-as-a-service-platform/</guid><description>A hands-on review of ModelScope, Alibaba DAMO Academy&apos;s open-source model hub. Covers SDK setup, model discovery, ms-swift fine-tuning, and how it compares to Hugging Face for Qwen-family and DAMO research workflows.</description><pubDate>Tue, 12 May 2026 08:18:10 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Paperless-ngx: Self-Hosted Document Management for Developers Who Want the API</title><link>https://pickuma.com/posts/paperless-ngx-self-hosted-document-management-for-developers/</link><guid isPermaLink="true">https://pickuma.com/posts/paperless-ngx-self-hosted-document-management-for-developers/</guid><description>A hands-on review of paperless-ngx, the open-source self-hosted DMS. We cover the Docker stack, OCR pipeline, REST API, AI workflow integration, and where Whoosh search hits its limits.</description><pubDate>Tue, 12 May 2026 08:16:54 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>Gitleaks: Open-Source Secret Scanning for Git Repos in 2026</title><link>https://pickuma.com/posts/gitleaks-open-source-secret-scanning-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/gitleaks-open-source-secret-scanning-2026/</guid><description>A hands-on review of Gitleaks for catching hardcoded secrets in Git history. Covers the CLI, pre-commit hooks, CI integration, and how it stacks up against GitGuardian for teams that don&apos;t want to pay per developer.</description><pubDate>Tue, 12 May 2026 08:15:24 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>Eleven Browser Games in a Week — What a Design System Buys You</title><link>https://pickuma.com/posts/eleven-browser-games-design-system/</link><guid isPermaLink="true">https://pickuma.com/posts/eleven-browser-games-design-system/</guid><description>play.pickuma.com now has eleven games. After the first two, the bottleneck was no longer game logic — it was the chrome around each game. Here&apos;s the design system that made the next nine fast.</description><pubDate>Tue, 12 May 2026 08:00:00 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>I Shipped Two Web Games This Weekend — Here&apos;s the Stack</title><link>https://pickuma.com/posts/two-web-games-weekend-build-launch/</link><guid isPermaLink="true">https://pickuma.com/posts/two-web-games-weekend-build-launch/</guid><description>Stop at 7.77 and Eagle Run are live at play.pickuma.com. A 250-line vanilla canvas game and a one-button time-sense test, both shipped in a day. The stack, the tradeoffs, the things that worked.</description><pubDate>Tue, 12 May 2026 07:15:00 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>AdamsReview: Multi-Agent PR Reviews for Claude Code, Reviewed</title><link>https://pickuma.com/posts/adamsreview-multi-agent-claude-code-pr-review/</link><guid isPermaLink="true">https://pickuma.com/posts/adamsreview-multi-agent-claude-code-pr-review/</guid><description>AdamsReview orchestrates multiple Claude Code agents for PR reviews. We break down how multi-agent review catches what single-pass LLM reviews miss, and where it fits in your pipeline.</description><pubDate>Tue, 12 May 2026 06:22:09 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>AI Note-Takers and Legal Risk: What Developers Should Know in 2026</title><link>https://pickuma.com/posts/ai-note-takers-legal-risk-developers-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/ai-note-takers-legal-risk-developers-2026/</guid><description>Otter, Fireflies, and Granola are facing class actions over consent and data retention. Here&apos;s what developers integrating AI transcription need to audit before shipping.</description><pubDate>Tue, 12 May 2026 06:20:35 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Claude as a User-Space IP Stack: What an ICMP Ping Benchmark Reveals About LLM Latency</title><link>https://pickuma.com/posts/claude-user-space-ip-stack-ping-latency-benchmark/</link><guid isPermaLink="true">https://pickuma.com/posts/claude-user-space-ip-stack-ping-latency-benchmark/</guid><description>Adam Dunkels wired Claude into a user-space TCP/IP stack and benchmarked it against ICMP ping. The latency floor it reveals is the most honest stress test we have for agentic Claude API workflows.</description><pubDate>Tue, 12 May 2026 06:19:22 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>yt-dlp: The CLI Video Downloader Developers Actually Use in 2026</title><link>https://pickuma.com/posts/yt-dlp-cli-video-downloader-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/yt-dlp-cli-video-downloader-2026/</guid><description>yt-dlp replaced youtube-dl as the default for programmatic video and audio extraction. Installation, format selectors, the Python API, and the production gotchas we hit running it across three real workflows.</description><pubDate>Tue, 12 May 2026 06:18:02 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Build Your Own X: 10 Project-Based Tutorials That Actually Teach You How Software Works</title><link>https://pickuma.com/posts/build-your-own-x-10-project-tutorials/</link><guid isPermaLink="true">https://pickuma.com/posts/build-your-own-x-10-project-tutorials/</guid><description>The build-your-own-x GitHub repo has 350k+ stars for a reason. Here are 10 from-scratch tutorials — databases, compilers, Git, neural nets — that teach how the tools you use every day actually work.</description><pubDate>Tue, 12 May 2026 06:15:49 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Phantom Pulse RAT Hits Obsidian Plugins: How to Audit Dev Tool Supply Chains</title><link>https://pickuma.com/posts/obsidian-plugin-phantom-pulse-rat-supply-chain/</link><guid isPermaLink="true">https://pickuma.com/posts/obsidian-plugin-phantom-pulse-rat-supply-chain/</guid><description>A malicious Obsidian community plugin delivered the Phantom Pulse RAT to developer vaults. Here is the attack chain and how to audit plugins in Obsidian, VS Code, and Cursor.</description><pubDate>Tue, 12 May 2026 06:13:07 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>Ratty Terminal Emulator: Inline 3D Graphics for Developers</title><link>https://pickuma.com/posts/ratty-terminal-emulator-inline-3d-graphics/</link><guid isPermaLink="true">https://pickuma.com/posts/ratty-terminal-emulator-inline-3d-graphics/</guid><description>A measured look at Ratty, a terminal emulator pitching inline 3D graphics. Where the category fits, which workflows benefit, and what to verify before you switch.</description><pubDate>Tue, 12 May 2026 06:11:21 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>AI Coding Agents Must Reduce Maintenance Costs, Not Just Write Code</title><link>https://pickuma.com/posts/ai-coding-agents-reduce-maintenance-costs/</link><guid isPermaLink="true">https://pickuma.com/posts/ai-coding-agents-reduce-maintenance-costs/</guid><description>Why evaluating Copilot, Cursor, and Claude Code by lines generated misses the point — and how to measure whether your AI tooling is adding or removing technical debt.</description><pubDate>Tue, 12 May 2026 06:10:01 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Audiorista vs Building Your Own Audio App: When No-Code Wins for Creators</title><link>https://pickuma.com/posts/audiorista-no-code-audio-app-vs-build-yourself/</link><guid isPermaLink="true">https://pickuma.com/posts/audiorista-no-code-audio-app-vs-build-yourself/</guid><description>A practical look at when a no-code platform like Audiorista beats spinning up your own audio app — and when it doesn&apos;t — for podcasters, course creators, and audiobook publishers who want to escape Spotify/Apple dependency.</description><pubDate>Tue, 12 May 2026 00:00:00 GMT</pubDate><category>saas-productivity</category><author>Owen</author></item><item><title>Signeasy vs DocuSign vs Dropbox Sign: Picking eSignature for an SMB SaaS</title><link>https://pickuma.com/posts/signeasy-vs-docusign-dropbox-sign-for-smb-saas/</link><guid isPermaLink="true">https://pickuma.com/posts/signeasy-vs-docusign-dropbox-sign-for-smb-saas/</guid><description>A practical breakdown of three eSignature platforms — Signeasy, DocuSign, and Dropbox Sign — for early-stage SaaS founders who need contracts signed without paying enterprise pricing or getting locked into Salesforce-only workflows.</description><pubDate>Tue, 12 May 2026 00:00:00 GMT</pubDate><category>saas-productivity</category><author>Owen</author></item><item><title>Woodpecker vs Lemlist vs Instantly: Cold Email Tools That Still Land in 2026</title><link>https://pickuma.com/posts/woodpecker-vs-lemlist-instantly-cold-email-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/woodpecker-vs-lemlist-instantly-cold-email-2026/</guid><description>After Google and Yahoo tightened sender requirements in 2024, half of the cold email tools that worked in 2023 are now dead in the water. Here&apos;s how Woodpecker, Lemlist, and Instantly hold up — and which one fits your team.</description><pubDate>Tue, 12 May 2026 00:00:00 GMT</pubDate><category>saas-productivity</category><author>Owen</author></item><item><title>Best Free Tiers for Developers in 2026: SaaS, PaaS &amp; IaaS Tools</title><link>https://pickuma.com/posts/best-free-tiers-developers-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/best-free-tiers-developers-2026/</guid><description>A 2026 audit of free-tier developer services: which hosting, database, CI/CD, and observability platforms still let you ship a side project for $0, where the hidden cliffs are, and when paying actually costs less than working around limits.</description><pubDate>Mon, 11 May 2026 23:29:23 GMT</pubDate><category>infrastructure</category><author>Owen</author></item><item><title>Mythos AI Found a Real Curl Vulnerability — What It Signals for Security Audits</title><link>https://pickuma.com/posts/mythos-ai-curl-vulnerability-security-auditing/</link><guid isPermaLink="true">https://pickuma.com/posts/mythos-ai-curl-vulnerability-security-auditing/</guid><description>Daniel Stenberg confirmed Mythos surfaced a real bug in curl, one of the most-reviewed codebases on the planet. Here&apos;s what that means for AI-assisted security review in your pipeline.</description><pubDate>Mon, 11 May 2026 23:27:50 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Running Local LLMs on M4 Mac with 24GB RAM: What Actually Fits</title><link>https://pickuma.com/posts/running-local-llms-m4-mac-24gb/</link><guid isPermaLink="true">https://pickuma.com/posts/running-local-llms-m4-mac-24gb/</guid><description>A measured guide to running 7B-32B local language models on a base M4 Mac with 24GB unified memory. Model size math, real tokens/sec numbers, and when Ollama, llama.cpp, or MLX is the right tool.</description><pubDate>Mon, 11 May 2026 23:26:29 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Why Developers Are Quietly Turning Off Copilot and Cursor</title><link>https://pickuma.com/posts/developers-ditching-ai-copilots-hand-coding/</link><guid isPermaLink="true">https://pickuma.com/posts/developers-ditching-ai-copilots-hand-coding/</guid><description>A measured look at the backlash against AI coding assistants — what the METR study and cognitive offloading research show about when hand-coding actually produces better engineers and better code.</description><pubDate>Mon, 11 May 2026 23:25:01 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Why Local AI Should Be the Default for Developers in 2026</title><link>https://pickuma.com/posts/local-ai-default-developers-2026/</link><guid isPermaLink="true">https://pickuma.com/posts/local-ai-default-developers-2026/</guid><description>The case for running models on your laptop instead of paying per-token API bills: where local AI (Ollama, LM Studio, llama.cpp) wins on cost, latency, and privacy, and where the cloud still earns its keep.</description><pubDate>Mon, 11 May 2026 23:23:25 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item><item><title>Cursor vs VS Code: We Ran Both for 30 Days</title><link>https://pickuma.com/posts/hello-cursor/</link><guid isPermaLink="true">https://pickuma.com/posts/hello-cursor/</guid><description>A practical 30-day comparison of Cursor and VS Code across multi-file edits, agent workflows, and pricing — based on actual usage.</description><pubDate>Mon, 11 May 2026 00:00:00 GMT</pubDate><category>ai-dev-tools</category><author>Owen</author></item></channel></rss>