Articles, tutorials, and thoughts on web development, tech, and building cool things.
What changed when I moved my engineering work onto Claude Code agents. Measured throughput, speed and quality, what I built, what broke, and what I'd tell another engineering leader.
SBI Research expects the RBI to hike, not cut. At SBI's own rates, a 5-year FD already pays less than a 2-year one. The math on waiting versus locking.
Revenue per employee rose 3-4% across India's top IT firms in FY26, after two flat decades. What actually caused it, and what FY27 has to prove.
TCS cut 23,000 jobs while Infosys hired 20,000 freshers. The real Indian IT AI workforce split is subtler, and FY27 will show whose bet pays off.
Three tools on this site shipped with real bugs: a calculator that lied, a leaked API key, a converter broken on every input. How I found and fixed them.
Agentic coding tools read your repo, run commands, and push code. Prompt injection, leaked secrets, supply-chain risk, and a pre-flight checklist.
Build a working MCP server in TypeScript: tools, resources, stdio and HTTP transports, and testing with a real client. Uses the v2 SDK, full code included.
AI token prices collapsed roughly 300x since 2023, yet compute spend keeps setting records. Why cheaper intelligence means bigger bills, not smaller ones.
GPUs stopped being the constraint on AI scaling. Grid queues, transformer lead times and cooling water took over, and India's buildout feels it first.
A build log from two browser games on this site — Phaser scene architecture, React lifecycle bugs AI could not fix, and what a deterministic sim taught me.
The Market Mood Index in my SIP/FD calculator runs on three inputs and a weighted average. Here is the exact arithmetic, and the four things it cannot see.
Reliable AI agents fail on plumbing, not intelligence. The real failure modes, the reliability stack that fixes them, and a 10-item production checklist.
Most SIP vs FD comparisons use pre-tax numbers. Taxing both correctly does not narrow the gap, it widens it. The worked math at 20% and 30% slabs.
The 2023 tax change killed indexation on debt funds. Here is how PPF, NSC, SCSS and SSY actually compare to debt funds now, using current GoI rates.
Agentic AI isn't killing IT services, it's killing the slack. Why margin compression is reshaping TCS, Infosys and Wipro, and the pivot that survives.
A Linux Foundation/Meta report finds 89% of AI users rely on open-source models, with 25% higher ROI. Comparing Llama, Mistral, and DeepSeek.
Indian IT and AI after Q1 FY27: TCS is hiring again, Wipro set no fresher target, HCLTech shrank. What the data means for your job and your next move.
EU AI Act deadlines after the Digital Omnibus: what applied on 2 Aug 2026, what lands in Dec 2026, Dec 2027 and Aug 2028, plus a developer checklist.
Waterfall, Agile, AI Build, Vibe Coding — AI promises a prototype in days. But most teams want a reliable car, not a supercar. How to pick your approach.
Single models hit diminishing returns. The next gains come from making top models critique each other: debate, judge, synthesize — with starter code.
Anthropic shipped Fable 5, a tier above Opus. Its pricing and constraints reveal a philosophy: the more capable the model, the more carefully it's gated.
Uber burned its 2026 AI budget by April; Microsoft cut Claude Code seats. Yet the gains are real. Is agentic coding a leap or an expensive trend?
Enterprises will spend $665B on AI in 2026, yet 73% can't prove a return. A practical governance framework to stop experiments becoming budget leaks.
AI coding feels faster, but one study found developers 19% slower while feeling 20% faster. How to measure real AI ROI with DORA and SPACE.
Open-source AI isn't free — it shifts cost from an API bill to your payroll. The real total cost of ownership, the break-even point, and how to choose.
Will AI replace developers? An Indian IT engineering manager maps which tasks AI automates, what hiring data shows, and the moves that keep you valuable.
India's first GenAI unicorn quietly paused foundation models and chips to become a cloud company. It looks like defeat — it may be the honest move.
Untangling Indian IT's manager alphabet soup — Project, Program, Product, Delivery, and Engineering Manager. What the EM role really is in the AI era.
A transparent, data-driven framework for measuring your AI displacement risk — 40+ roles including India-specific ones, refreshed quarterly.
Install Claude Code in 5 minutes, then build your first project with plan mode, CLAUDE.md and the commands that matter. Checked against October 2026 docs.
From zero to 55+ articles, 7 interactive tools, and 280+ pages — here's how I used Claude Code and AI to build my personal website in weeks, not months.
Master prompt engineering: the 6 elements of effective prompts, mistakes to avoid, and advanced techniques that work with ChatGPT, Claude, and Gemini.
Anthropic's Claude Mythos found thousands of zero-days and wrote browser exploits autonomously — then chose not to release it. An EM's take.
Google Stitch's new design-markdown export looks minor but isn't. Why consistent AI-generated design across projects matters more than it appears.
AI design tools are changing UX work — but not how most think. The role isn't disappearing; it's evolving into something that demands new skills.
Tailwind laid off most of its team in 2026. Is this the end of CSS frameworks? An honest look at what AI means for frontend — and what to learn next.
Google's AI tool Stitch now turns vibes into interactive prototypes in seconds. What changed, why it matters, and what developers should do about it.
India faces rising high-skilled emigration. Instead of despair, the honest data: why it's happening and what it means for those who stay and build.
32 million Indians live abroad, sending home $100B+ a year. But the real story is the networks, knowledge, and capital flowing back to build India.
Voice prompts, URL ingestion, design-to-prototype in seconds — vibe coding isn't a gimmick. A practical guide to using it without losing control.
Some devs have AI superpowers; others are locked out. Stakeholders expect 10x, costs balloon, the SDLC persists. What's really happening in AI delivery.
219,000 GitHub stars, a craze in China, and it runs on your own machine. OpenClaw is quietly changing how developers work — and India's missing it.
The junior developer job market has fundamentally changed. What every 2026 graduate needs to survive and thrive in an AI-augmented workplace.
AI could add $1 trillion to ASEAN economies — or widen the rich-poor gap for good. Why 2025-2027 decisions set 50 years of economic outcomes.
A billion working-age Indians: in the AI age, either a superpower or an economic disaster. Why the 2025-2027 window sets India's next 50 years.
$285 billion vanished from SaaS stocks in a day, and Indian IT firms are poised to seize the chaos. Why this may be India's biggest opening since Y2K.
We're told anyone can learn anything with effort. But what if we've misjudged human potential? A provocative look through the lens of AI architecture.
The science of caffeine, developer coffee culture, brewing schedules, and why 4 AM bug fixes need a steaming mug. A guide to caffeinated coding.
Explore hilarious, brutal, and honest commit messages from developers worldwide. A celebration of code, chaos, and the raw truth behind the screen.
The science behind rubber duck debugging, why it works so well, and how to use the technique in your workflow — with real stories and setup tips.
How AI sessions lose memory through context compaction, why it matters, and what Leonard Shelby teaches us about working with forgetful assistants.
Many developers treat AI as temporary hype and wait for it to 'stabilize.' Why that mindset costs careers, with examples from India's IT industry.
A practical AI freelancing guide for Indian developers: winning profiles, proposal templates, pricing, and case studies of $5K-15K/month earners.
AI jobs in India, minus the hype: the skills that pay, what ML and GenAI engineers actually earn, and which companies are hiring, from an engineering lead.
The 20 best AI platforms, refreshed for September 2026: Claude Fable 5.1, GPT-6 Astra, Gemini 3.8 Flash, Grok 4.7, current prices and picks by use case.
Complete tutorial to build your first AI agent using free tools. Step-by-step guide with full code to create a Personal Research Assistant in 2-3 hours.
Anthropic now includes Claude Code on every Team plan seat. Why democratizing elite AI coding tools will reshape enterprise software in 2026.
Grok vs Claude vs ChatGPT vs Gemini for developers: hands-on coding tests, India pricing and payments, plus the September 2026 models and API rates.
Escape the service mindset trap. Learn how Indian developers are building AI products with low-cost tools, real success stories, and a 30-day MVP roadmap.
How Meta's Vision-Language Joint Embedding Architecture (VLJA) challenges modern AI by predicting meaning instead of tokens — a possible post-LLM era.
Model Context Protocol went from a Nov 2024 release to 97M SDK downloads, adopted by ChatGPT, Gemini, and VS Code. A technical getting-started guide.
India's 1,700+ Global Capability Centers are shifting from back-office to strategic innovation hubs — heading toward a $90 billion market by 2028.
Microsoft, Google, AWS, and Indian giants are pouring $50B+ into data centers. India's AI infrastructure buildout, power challenges, and opportunities.
100% of Claude Code's recent codebase was written by Claude Code itself. From Shopify to Google, leaders are stunned. Why you may already be behind.
Copilot vs Cursor vs Windsurf (now Devin Desktop) vs Claude Code, updated for September 2026: pricing, default models, agent features, and which to pick.
Why 'good enough' AI beats bleeding-edge for most use cases — the shift from hype to practical ROI, and what actually drives enterprise value.
Apple's LLM-powered Siri slipped to September 2026 and now leans on Google's Gemini. Is Apple's privacy-first restraint still an advantage?
Why 2025's agentic AI pilots failed, what's different now, and how to build production-ready AI agents that survive real enterprise workloads.
A realistic look at India's place in the global AI race — data sovereignty, cloud dependency, and the path to genuine tech independence.
A practical guide to LoRA and RAG — two ways to enhance large language models. When to use each, and how to combine them for the best results.