Mumbai, India
For ten years, I've shipped software in production. For the last two, I've been pulling apart Claude Code, MCP, and agent frameworks to understand what's underneath the demos. I share the working pieces on YouTube, including the parts that broke, which is usually more useful.
10+
Years shipping production software
PromptMate
Live on the Chrome Web Store
5-Check
Framework for AI-generated code
How I work
I put AI tools into normal engineering workflows, watch where they help, and document the judgment needed around them.
01
I use Claude Code, MCP servers, and agent workflows on concrete software problems before turning them into lessons.
02
The useful parts usually sit next to the failures, wrong turns, and missing context. I show both.
03
When something repeats, I shape it into a build, checklist, video, or framework other engineers can reuse.
Latest videos
AI tools, engineering workflows, and lessons from building in production. Videos in Hinglish.
How OpenClaw actually works internally, why it burns 10,000 tokens before your first message, and what makes it fundamentally different from Claude Code or Cowork. Real use cases, honest security risks, and whether you should try it today.
Watch
How Claude Cowork automates the entire workflow (reading websites, analyzing content, and delivering summaries directly to Slack) without a single line of code. The Brain, Eyes, Mouth mental model for AI automation.
Watch
One terminal, one agent, 15-minute wait per task. What if you ran 3 AI agents in parallel on the same codebase? Production-grade workflow for 3-4x development speed without sacrificing code quality or safety.
WatchOpen source
Tools I built to solve real problems. All open source, built in public.
An MCP server that connects Content Board (my Firebase PWA for YouTube production) to Cowork, Claude's desktop agent. I built it live on the channel, episode by episode, so the whole thing is documented end to end. It's the working proof that the MCP tutorials on this site come from a server I actually ship and use every week.
Chrome extension that adds a personal prompt library sidebar to ChatGPT and Claude. Save prompts with tone and format settings, insert them in one click, and sync across devices via Google Drive. No backend server.
A folder-based workflow automation system for Claude Cowork. Define repeatable AI workflows once: manual pipelines with approval gates, scheduled tasks, multi-stage projects. Includes a YouTube content pipeline and daily community scout out of the box.
Frameworks
The videos show the experiments. The frameworks capture the repeatable parts: task boundaries, review commands, security checks, manual testing, changeset review, and human judgment.
Explore frameworksCurrent framework
Trust, verify, override before merge
Engineering background
Before I started publishing AI workflows, I spent 10 years building enterprise systems. That background is why I know when AI helps and when it does not.
Today I'd use AI to automate the tenant-by-tenant migration validation. Back then, we did it manually.
View case study
This was AI in production before the hype cycle. Probabilistic models making real deployment decisions.
View case study
Legacy modernization at its messiest. The kind of system complexity where AI-generated code breaks on day one if you don't understand the domain.
View case study
Identity systems have zero margin for error. AI accelerates the boring parts, but a human must own the critical path.
View case studyFollow the work
Start wherever you think best: watch the experiments, inspect the artifacts, or reuse the checks in your own workflow.
Videos
Claude Code, MCP servers, and agent workflows tested in real engineering contexts.
OpenBuilds
Open source projects that came from solving actual workflow problems.
OpenFrameworks
Practical checklists and mental models shaped from repeated AI tool work.
Open