nitishagar

From progressive tool discovery to the new three-layer architecture, here’s how the Model Context Protocol evolved for 2026.

Is AI writing tests that protect bugs? Why TDD and Hexagonal Architecture are the only ways to ground AI-generated code.

I built SoberAI because I kept seeing the same problem: AI crawlers (GPTBot, ClaudeBot, PerplexityBot) don't execute JavaScript. Your React SPA is literally a blank page to them. There was no tool to audit for this — so I made one. SoberAI is an Electron app that scores your site 0-100 across four categories: SSR readiness, Schema.org coverage, semantic HTML structure, and content extractability. It renders your page twice with Playwright (JS on vs off) and diffs the content. LLM-powered recommendations are optional (Ollama local or OpenAI). One command: npm install && npm run electron:dev No accounts. No telemetry. MIT licensed. ~3K lines of JS. Docs: https://nitishagar.github.io/sober-ai Would love feedback on the scoring methodology and what auditors to add next.

I wrote about "Agentic Debt" and the new OWASP LLM07:2025 classification. The industry is finally converging on a standard: Prompts are not secrets or config.

Slash incident response time by 40%. Explore multi-agent patterns and RAG strategies used by Netflix, Uber, and Microsoft to automate Ops.

LLM agents break deterministic tests. Evaluation-Driven Design helps teams reduce hallucinations.

The AI revolution just flipped: small AI models could redefine impact while using 70x LESS energy and costing 1,000x LESS to deploy.

In my new article I explore why compact AI might be the future.

"Orchestration Outperforms Monolithic Agents"

I try to dive into how Salesforce BOLAA system can help you build efficient and reliable systems cheaper.

The myth of the 10x engineer is becoming a reality, but the engineer isn’t human. It’s an AI agent, and your job is to become its conductor.

Building in AI? Decide if you're shipping the "picks & shovels" (MCP server platform) or the "gold" (AI Agent) – your strategy depends on it! #AI #ProductStrategy #MCP #AIAgents #TechInnovation

Discover how Agentic AI is shifting digital accessibility from reactive fixes to proactive, built-in inclusion for everyone.

In this recent article, I explore the contrasting approaches of core AI technology providers, from the broad capabilities of foundation models to the precision of vertical AI solutions.

The article examines how DeepSeek serves as a catalyst for companies to explore cost disruption through AI tools and augmentation.

My latest article explores the potential of TPUs, NPUs & more, and whether you should be looking beyond the cloud. #AI #Hardware #Innovation #Tech #ArtificialIntelligence

I just published story on how AI content boom demands ethical approaches – have a read! https://link.medium.com/nCMmoZj08Pb #AI #EthicsInAI #TechEthics #GenerativeAI #AIContent

My take on building killer customer support AI: ditch the complexity, embrace smart simplicity! Sharing my insights after diving into the latest strategies for #LLMs and #ConversationalAI in #CustomerSupport.

Thoughts on state of security in customer support and why it needs a makeover now.

#APIGovernance #AIIntegration #EngineeringLeadership #APIDevelopment #AIinAPIs

This piece explores the impact of artificial intelligence on API governance and examines potential AI applications for enhancing the developer experience.

The future of AI isn’t about building bigger models; it’s about improving the quality of data. In the article below, I highlight key areas to explore beyond just model size.

Gen AI is a powerful tool that can help businesses provide better customer service at scale. By automating tasks, providing real-time support, and personalizing interactions, Gen AI can help businesses improve efficiency, increase productivity, and improve customer satisfaction.

While other industry players have certainly seen significant advancements, AI stands out as being on an accelerating trajectory fueled by a potent combination of big data, computing power, and algorithmic breakthroughs.

Prioritize accessibility, leveraging the power of AI, helps us pave the way for a more inclusive digital future, where every user can experience the full potential of mobile apps, regardless of their abilities.

TL;DR: To unlock the true potential of LLMs for business success, augment them with domain-specific knowledge using techniques like few-shot prompting, fine-tuning, bootstrapping, reinforcement learning, and knowledge graph integration. Combine LLM reasoning with structured data to drive innovation and gain a competitive edge.