OpenAI's GPT-5.4 crosses the human baseline on OSWorld-V with native computer-use, 1M token context, and parallel tool calling. Here's what it actually means for teams building AI systems in 2026.
An 8B model that fits in 1GB RAM, runs on your iPhone, and stays competitive with full-precision models. PrismML's Bonsai changes the math on edge AI deployment. Here's the technical breakdown.
Stripe's autonomous agents write 1,300 PRs per week. OpenAI extended its Responses API with shell tools. AWS launched Strands Labs. The agentic shift is no longer theoretical — here's what it looks like in the real world.
Meta dropped Llama 4 Scout, Maverick, and Behemoth. Google fired back with Gemma 4. As a Technical Lead, here's what these releases actually mean for your teams and projects.
Alibaba just released its third proprietary model in days. Google's Gemini Flash-Lite costs $0.25 per million tokens. NVIDIA's Nemotron runs 2.2x faster than GPT-OSS-120B. The LLM cost war has arrived — here's what it means for architects choosing AI infrastructure in 2026.
The A2A protocol under the Linux Foundation is quietly becoming the HTTP of the agentic era. Here's what it means for enterprise architects, why it matters more than another model release, and how to think about it from a systems design perspective.
GPT-5.4 crossed human performance on desktop tasks. With a hosted shell, 1M context window, and native compaction, the era of AI as an autonomous coworker is finally here — and the implications for developers are profound.
This week's top reads for Tech Leads & Engineers: the AI coding tools war heats up, Stripe's autonomous agents ship 1,300 PRs/week, software supply chain becomes the #1 attack vector, and engineering leadership rethinks itself.
Anthropic acquired Bun in December 2025. OpenAI acquired Astral (uv, Ruff) in March 2026. The AI model war has a new front: owning the developer toolchain. Here's what this arms race means for how you build software.
Apple just committed $1 billion per year to run Google Gemini inside Siri. As a Technical Lead who builds on Apple's platform, here's what I think this means beyond the headlines — privacy tradeoffs, WWDC expectations, and what APIs you should be watching.
Cursor's March 2026 release brings parallel subagents, BugBot Autofix, and cloud agents — a genuine shift from AI coding assistant to AI workflow engine. Here's what it means for real teams.
The CNCF just made Dapr Agents v1.0 generally available. After years of AI agent frameworks focused on intelligence, here's one that focuses on survival — durable workflows, state management, and production-grade reliability on Kubernetes.
Google just made Gemini Code Assist free and priced Flash-Lite at $0.25/M tokens. After 15 years building production systems, here's what this cost collapse really changes about how we build.
The Model Context Protocol crossed 10,000 published servers under the Linux Foundation's Agentic AI Foundation. As someone who's integrated dozens of AI systems, here's why this number matters more than any benchmark.
GPT-4 level AI cost $30/M tokens in 2023. Today it's under $1. Here's the technical architecture that lets you capture 90%+ of that savings without sacrificing quality.
Mistral's Devstral 2 achieves 72.2% on SWE-Bench at 7x cheaper than Claude Sonnet. Here's what this means for teams building AI-powered dev tooling in 2026.
A comprehensive guide to the modern free-tier tech stack that lets you build, deploy, and scale a startup for roughly $20/month. No servers. No DevOps team. No funding required. Just an idea and WiFi.
90% of developers now use AI at work. But the real shift in March 2026 is agents moving from suggestion-mode to autonomous execution. Here's what that actually looks like in production systems and what breaks when you go too far too fast.
Deep dive into Gemini 3.1 Pro's practical developer features — 77.1% ARC-AGI-2, dynamic thinking API, 1M token context, and what these benchmarks actually mean when building production AI systems.
March 2026 marks a turning point — AI coding tools evolved from assistants to autonomous agents. As a Technical Lead, here's what you actually need to know to navigate JetBrains Central, GitHub Agent HQ, Dapr Agents, MCP, and the multi-agent revolution.
Beyond benchmarks — a real comparison of DeepSeek V3.2 and Gemini 3 Pro for production developer workflows. Cost, code quality, context window, multimodal capabilities, and when to use which.
Claude Opus 4.6, GPT-5.4 Thinking, and Gemini 3.1 Pro all shipped within the same month. Here's a practical guide to choosing the right model for your team — based on real benchmarks and production patterns, not marketing slides.
Gartner says 40% of enterprise apps will embed AI agents this year. But 40% of agentic projects will be scrapped by 2027. Here's what separates the teams that ship production agents from those that get stuck in pilots forever.
Mistral's Devstral 2 scores 72.2% on SWE-bench, ships under MIT license, and costs up to 7x less than Claude Sonnet. Here's how it works, when to use it, and whether open-source agentic coding is finally production-ready.
Anthropic's Agent Teams feature in Claude Opus 4.6 lets multiple Claude Code instances work in parallel on the same codebase. Here's the architectural model, real-world performance data, and what actually changes for teams building production software.
GPT-5.4 scored 75% on OSWorld-Verified, surpassing human baseline of 72.4%. Here's what this means for developers building agentic systems in 2026, with real API examples and architectural patterns.
How AI is reshaping software architecture from the inside — ML-powered pattern selection, LLM orchestration, AI gateways, event-driven agents, and a full team implementation roadmap with diagrams and production-ready code.
Anthropic refused DoD's mass surveillance deal. OpenAI signed it and lost millions of users overnight. This is not an abstract ethics debate — it's a supply chain risk that every technical lead building on AI APIs needs to understand now.
GPT-5.4, Gemini 3.1 Pro, and Claude 4.6 are now neck-and-neck on benchmarks. When the models are equal, everything else becomes the differentiator. Here's how to choose.
Product owners think in outcomes. Developers think in systems. AI can bridge that gap — with the right documents, workflows, and prompts. Here's exactly how.
Code generation gets all the hype, but AI's biggest impact might be in the 80% of development that isn't writing code: requirements, docs, architecture decisions, and communication.
The series covered how to prevent bugs. This post covers what to do when they slip through. How to use AI as a debugging partner without chasing its hallucinations.
AI works differently for every developer. How to establish team conventions, shared context, consistent quality, and avoid the 'five different architectures' problem.
How to integrate AI assistants into your Umbraco migration workflow: prompt patterns for Umbraco-specific tasks, using Umbraco 17's built-in Developer MCP, Architectural Decision Records (ADRs) as migration governance, and the AI Champion model for tech teams.
The series covered WHAT to do with AI. This post covers HOW to communicate with it. Prompt patterns that produce usable code, not impressive demos.
AI doesn't understand threat models. How to define your trust boundary, handle security-critical code, and build guardrails that protect your users.
When you're managing 10, 20, or 50 Umbraco sites, individual project economics don't work. The Marketing OS framework: shared NuGet packages, shared document type libraries, AI-accelerated delivery, and how to reduce per-site migration cost by 50–70%.
AI-generated code needs different testing strategies. How to write tests that catch what AI gets wrong, use AI to generate test cases, and build a testing workflow for AI-assisted development.
Everything we learned building BuildRight with AI, the mistakes we'd make again, the ones we wouldn't, and where AI-assisted development is heading.
The final retrospective. Ten anti-patterns to avoid, what AI genuinely can't do, what surprised us, and where AI-assisted migration is heading next.
AI can generate impressive code that completely ignores your architecture. How to provide architectural context, enforce patterns, and know when to override AI suggestions.
Gut feelings aren't metrics. How to measure AI's real impact on speed, quality, and team satisfaction — with honest numbers from the BuildRight project.
5 detailed case studies showing how different teams implemented AI workflows with Claude Skills, Gemini Gems, and NotebookLM. Includes step-by-step playbooks, metrics, and lessons learned.
The part nobody talks about: teaching a junior developer to use AI effectively, managing team resistance, earning client trust, and keeping the boss updated. Migration is as much people as it is code.
Practical AI-powered workflows for every Agile ceremony: sprint planning, daily standups, retrospectives, backlog refinement. Includes skill templates and real-world implementation guide.
Most developers skip straight to 'generate code.' The teams that get real value from AI spend 40% of their time planning before writing a single prompt.
How to build, maintain, and govern shared AI skill libraries for teams. Covers skill catalogs, onboarding workflows, quality gates, and practical governance frameworks.
The definitive guide to Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) in 2026. Learn how to get your content cited by ChatGPT, appear in Google AI Overviews, and be recommended by Perplexity and Claude. Part 8 of the SEO Leader's Complete Playbook.
Before you write a single line of migration code, you need to know what you have. How to use AI to audit a legacy Umbraco project, score complexity, estimate effort, and produce the assessment document that drives every decision that follows.
Detailed comparison of Claude Skills vs Gemini Gems vs NotebookLM, plus 4 powerful combo workflows that chain all three tools together for maximum output.
Full blueprint for building a web app that generates daily English lessons, tracks your progress, and unlocks the next day's content only after completion. Includes Next.js code, AI integration, and assessment system.
Advanced guide to building AI workflow systems at scale: skill libraries, team governance, quality assurance, security compliance, and ROI measurement. For team leads and enterprise architects.
5 proven prompt patterns for generating high-quality Playwright tests with AI. Learn to control AI output, review generated code, avoid common mistakes, and build reusable prompt templates. Part 7 of the QC team's guide to automation with AI.
The review phase is 40% of the workflow for a reason. A real story about shipping too fast, the review checklist that prevents it, and how to build review habits that scale.
The complete AI-Powered Software Teams playbook: the full delivery pipeline, team configuration principles, toolstack recommendations, the 5 non-negotiable human checkpoints, and what it means to lead and work in an AI-augmented software team.
4 complete AI workflows tailored for business owners, developers, writers, and content creators. Step-by-step playbooks using Claude Skills, Gemini Gems, and NotebookLM together.
A hands-on comparison of three AI tools for test automation. See how Claude Code with Playwright MCP, GitHub Copilot, and Antigravity each approach the same test scenario, with setup guides and decision matrix. Part 6 of the QC team's guide to automation with AI.
How AI transforms every team ceremony: async-first standups with AI summaries, AI-prepared refinements, structured retrospectives with pattern detection, and sprint reviews where AI generates the demo context. Plus the human connections that meetings are really about.
The definitive guide to Google NotebookLM: Audio Overviews, Mind Maps, Deep Research, and 5 practical workflows for business owners, developers, writers, and content creators.
A complete technical blueprint for building a parent-child English learning application. Covers React Native/Expo mobile and Next.js web implementations, AI integration, and gamification.
Putting AI into production is nothing like building a demo. Two tech leads discuss costs, hallucinations, latency, guard rails, and what actually breaks when real users hit your AI features.
Umbraco 8 is already end-of-life. Umbraco 13 expires in December 2026. Here's the real cost of staying put, what Umbraco 17 delivers, and how AI changes the economics of every migration you've been delaying.
The real cost of AI voice interviews, broken down per minute. Managed vs self-hosted economics, the three tipping points, and how to get from $3.45 per interview to under $1.00.
How AI transforms the DevOps/Platform Engineer role: AI-assisted pipeline generation, incident diagnosis, capacity planning, GitOps automation, and the operational judgment that keeps systems running and teams unblocked.
A five-phase workflow for AI-assisted legacy migration. From assessment to ship, with tool recommendations, team role definitions, and the prompt patterns that actually work.
Master Gemini Gems: the Four Pillars framework, Knowledge Files strategy, 5 production-ready Gems for business, marketing, dev, and creative work. Step-by-step instructions.
How to use Claude, ChatGPT, and other AI tools as your personal English tutor. Prompt templates, role-play scenarios, and daily AI workflows for busy professionals.
Machine learning, LLMs, embeddings, and RAG — explained without hype. Two tech leads break down what AI actually does, using analogies about libraries, pattern matching, and really smart parrots.
Integrating Google Gemini API with Umbraco for AI-powered content generation, multi-language translation, SEO meta optimization, and a content review workflow that keeps humans in the loop.
A step-by-step walkthrough of building a landing page with AI assistance, demonstrating the 40-40-20 workflow in the simplest possible context.
How AI transforms the Security Engineer role: AI-assisted threat modelling, automated SAST/DAST pipelines, dependency scanning, security policy generation, and the human security judgment that keeps systems and users safe.
Everything you need to know about Claude Skills: structure, the P-T-C-F framework, 5 ready-to-use skills for developers, marketers, and writers. Copy-paste code included.
Recording interviews is easy. Doing it legally isn't. LiveKit Egress, real-time transcription, GDPR consent flows, HIPAA encryption requirements, and the compliance checklist you need before go-live.
Two migration projects, a small team, a cost-conscious boss, and AI as the equalizer. Here's what the data says — and what nobody tells you about migrating legacy systems in 2025.
What is AI hallucination? Why does AI invent facts? Learn detailed strategies for minimizing hallucinations in Claude, Gemini, Copilot, and Cursor — from prompt engineering to RAG, grounding, and verification loops.
How AI transforms the Technical Architect role: AI-assisted IaC generation, cost modelling, capacity planning, non-functional requirements analysis, disaster recovery planning, and the platform judgment that only experience provides.
A comprehensive guide to the 3 most powerful AI workflow tools available today. Whether you're a business owner, developer, writer, or content creator — here's how to work with AI 10x more effectively.
The complete guide to SEO in 2026. From Google's AI Overviews to E-E-A-T, zero-click searches to multi-platform visibility — everything SEO teams need to know about ranking in the age of AI. Part 1 of the SEO Leader's Complete Playbook.
Adding video analysis to voice interviews with Gemini Live's multimodal capabilities. Frame sampling, parallel processing, screen sharing for coding interviews, and keeping costs under $0.40 per session.
AI coding tools promise 10x productivity, but most teams see 1.2x. The difference isn't the tool. It's the workflow.
How AI transforms the QA/QC role: AI-generated test cases, exploratory testing agents, quality gate automation, defect prediction, and the human testing judgment that keeps products safe to ship.
Your voice interviewer is only as good as what it knows. Here's how to build a real-time RAG system that retrieves rubrics, job descriptions, and technical knowledge in under 50ms.
How AI transforms the Tech Lead role: AI-assisted PR review, standards enforcement, technical debt scoring, architectural governance, and the mentoring and culture-building that only great engineers provide.
Practical GitHub Copilot setup for an ecommerce development team. Workspace instructions, custom instructions per task type, Copilot Chat for architecture, and the MCP server for Angular CLI.
A hands-on guide for researchers, students, and executives — how to use Claude Web, Desktop, and Code to research any topic, write reports, create slides, and generate spreadsheets. No coding required.
Three personas, one infrastructure. How to build an AI interviewer that asks questions, a coach that gives feedback, and an evaluator that scores fairly — all with system prompts, function calling, and state machines.
How AI transforms the Solution Architect role: AI-assisted architecture diagramming, ADR generation, trade-off analysis, technology selection rationale, and the architectural taste that only experience provides.
Plan-before-code workflow, subagents for parallel tasks, context management strategies, debugging with Claude, the anti-patterns that waste time, and honest lessons from 6 months of daily production use.
Deepgram Nova-3 or Whisper for STT? Gemini Flash or GPT-4o for conversation? ElevenLabs or Cartesia for TTS? Real benchmarks, real costs, and the combinations that actually work in production.
How AI transforms the PO and PM roles: AI-assisted backlog scoring, sprint planning, status reporting, risk flagging, release note generation, and stakeholder communication — and the human judgment that keeps it grounded.
Writing custom skills with SKILL.md, slash commands, the official and community skill repos, GitHub PR automation with @claude mentions, GitHub Actions integration, and team workflow patterns.
How AI agents transform the BA role: from manual interview synthesis to structured requirements in hours, AI-assisted user story generation, acceptance criteria validation, and the human skills that stay irreplaceable.
What MCP is, how to configure servers in .mcp.json, the top servers by category (GitHub, Playwright, databases, search, design, cloud), performance pitfalls, and why you should limit yourself to 2-3 per project.
The cascaded STT→LLM→TTS pipeline gives you control. Speech-to-speech models give you speed. Here's how to choose — and why the best systems use both.
A panoramic view of how AI agents transform every role in a modern software delivery team: BA, PO, PM, SA, Tech Lead, Dev, QA, TA, Security, and DevOps — and what it means for how teams collaborate, decide, and ship.
VS Code extension setup, Command Palette workflows, inline diffs, @-mentions for file context, multi-tab conversations, auto-accept mode, Extended Thinking, and the daily habits that actually stick.
The landscape of real-time voice AI has shifted. Gemini Live, OpenAI Realtime, Bedrock Nova Sonic, and Grok make sub-500ms AI conversations possible. Here's the reference architecture for building a production voice interview platform.
Installing Claude Code, writing an effective CLAUDE.md, the /init command, memory architecture, .claude/ directory structure, and turning a generic AI into a project-aware pair programmer.
A complete reference guide to Cloudflare's free and low-cost services for three use cases: self-hosting homelab services, building full-stack web apps on the edge, and running AI workloads. Pricing tables, free tier limits, and use-case mappings — all in one place.
How to test LLM outputs, validate RAG retrieval quality, and verify vector search accuracy using DeepEval, Ragas, and Testcontainers. Part 2 of the AI-powered quality engineering series.
Setting up Playwright MCP with Claude Code to build an AI-powered test automation framework for a Next.js app. Complete guide covering Page Object Model, custom fixtures, network mocking, CI/CD integration, debugging, and AI prompting best practices. Part 1 of a series on quality engineering in the AI era.
A complete guide to connecting Jira, Linear, GitHub, and Playwright MCP servers with Claude Code to auto-generate BDD tests from ticket acceptance criteria, execute them against a live app, and report results back.
How I designed, analyzed, implemented, and tested Kids Learn — an AI-powered educational SaaS platform — using Claude as my development partner, Gemini for AI features, Next.js, PostgreSQL, and pgvector. A complete walkthrough from napkin sketch to production.
Lessons from building a real-time AI interviewer with LiveKit, OpenAI Realtime, Gemini Live, and Bedrock Nova. VAD tuning, provider failover, latency budgets, and the things nobody warns you about.
How I set up Moltbot — the open-source personal AI assistant — to automate code reviews, deployments, monitoring alerts, and daily standups. A practical guide for developers who want AI that actually does things.
I've used Claude Code, GitHub Copilot, and other AI tools every day for over a year. Here's what actually saves time, what wastes it, and the productivity trap nobody talks about.
I built a GitHub Action that uses Claude to review PRs. The first version was annoying. The current version catches real bugs. Here's the whole journey.
I experimented with CrewAI and LangGraph to build multi-agent workflows. Some were genuinely useful, some were expensive toys. Here's what I learned.
We built a RAG knowledge base. The first version gave wrong answers half the time. Six months of iteration later, it actually works. Here's every lesson.
We tested ChromaDB, Pinecone, Weaviate, Qdrant, and pgvector for our RAG system. Benchmarks, costs, developer experience — and the one I'd pick today.
We integrated Azure OpenAI into a client's enterprise platform. Here's the practical stuff — networking, content filtering, cost surprises, and what the docs don't cover.
I started with ASP.NET Web Forms in 2008. Now I'm building AI agents with .NET and Python. A reflection on how the ecosystem changed — and how it didn't.
Get notified when I publish new posts on AI, .NET, cloud architecture and more.
