About
Arjav Jain.
AI, Ops and Technology. COO at Hyperke, where I build outbound systems that handle 2-5 million cold emails a month. I teach what I learn at @lifeofarjav.
Based in India. Engineering background, operations mindset, builder.
The story
From engineering school to running outbound at scale.
I went to Shiv Nadar University for engineering. The program taught me how to understand systems before trying to fix them. That approach stuck with me.
Today I run operations and engineering at Hyperke. We're an outbound demand generation agency — we help B2B companies get more qualified leads, demos, and sales appointments. Our team sends 5-10 million cold emails a month. My job is making sure all that actually works.
Here's what I've noticed: most outbound tools look great until you use them with real data. Then they break. Bad company names, messy domains, duplicates — the whole thing falls apart. So I build the systems underneath: data cleaning, routing, deduplication, enrichment. The boring stuff that lets you send millions of emails without everything breaking.
On @lifeofarjav, I share the AI workflows and Claude Code setups that actually work. No theory, just the practical stuff I use every day. I also run Clay Club workshops, teaching revenue teams how to actually use the tools they're already paying for.
I write because it helps me think things through. If other people find it useful, that's a bonus. I'd rather ship something rough and useful than wait for perfect and never publish.
Timeline
Roughly how it went.
Origin
Shiv Nadar University
Engineering school, systems-first training. I learned to read how a system works before trying to fix it, and that habit never left.
Hyperke
Running outbound at scale
COO at Hyperke. Our team sends 5-10 million cold emails a month for B2B companies. I build the systems that make that possible — the routing, data cleaning, enrichment, all of it.
AI & Automation
Building the tech stack
I design the AI tools and automations that our team and clients use. Mostly the unglamorous engineering that keeps everything running when real-world data hits the system.
Teaching
Making AI practical
Through @lifeofarjav and Clay Club, I share the AI workflows and engineering fundamentals that actually work. The stuff people reference but never explain properly.
Today
Still building, still sharing
Still running the systems, still shipping the next one, still teaching what I learn. This site is where it gets written up.
What I believe
Three operating principles.
Build the layer underneath first
A marketing setup that demos well and dies on real data is worse than no setup, because it hides the failure until launch day. I'd rather ship something ugly that survives a messy company name than something polished that breaks on the first bad domain. The routing, the dedupe, the enrichment come before the landing page.
Explain hard things in plain words
Model Context Protocol, tool calling, cross-validation, agent loops. None of this needs jargon to explain, and dressing it up in jargon is usually a sign someone doesn't understand it yet. The test I use: if a reader can't repeat the idea back after one post, the post didn't work, and that's on me, not them.
Publish rough, publish often
If I waited until a piece was polished I'd publish almost nothing, and the version that ships at 80% teaches me more than the version that sits in drafts at 100%. Most of what I put out is written while I'm still figuring the thing out. That's usually where the useful parts are.
Work with me.
Two paths: outbound systems via Hyperke, or speaking / collabs / content directly.