Skip to content

Boss!

Co-founder, marketing and growth, 2026.

An AI recruiting startup in Bangalore.

Rudra co-founded Boss!, an AI recruiting startup, and ran all of its marketing, B2B and B2C. He built the go-to-market from zero: a brand system, website specs and design mocks, a researched outbound motion and a social workflow run by AI agents with a human approval gate.

What he built, at a glance

  • Social workflow

    • Scout agent
    • Writer agent
    • Verifier agent
    • Human approval
  • Brand system

    • Mascot library
    • Banner build script
    • Safe-zone checks
    • Platform-exact sizes
  • Website

    • Recruiter site spec
    • Design mocks
    • Competitor teardown
  • Research

    • Bangalore demand map
    • Job description corpus
    • Persona footprints
  • Outbound

    • Named decision-makers
    • Funded companies
    • Tailored pitches

The situation

Hiring is broken in both directions. Companies drown in applications and candidates disappear into them. Boss! set out to fix the first round of hiring with AI, and needed a market built from nothing.

What Rudra built

Social ran as a three-step workflow. A scout agent found conversations worth joining, a writer drafted the posts and replies, and a verifier blocked anything off-voice, unverified or against the disclosure rules before a person approved the queue.

He built the brand system, including an art-directed mascot library and a banner build script that checks every element against each platform’s safe zones before it exports.

He wrote the website specs and design mocks for the recruiter site after a teardown of competing recruiting sites.

The positioning came from research: a Bangalore hiring-demand map and a close read of real job descriptions across sales, software and marketing roles.

Outbound targeted named decision-makers at funded companies, with a tailored pitch for each one.

How the work ran

Most of the execution ran through agent workflows that Rudra designed and directed: research, drafting and adversarial verification split across agents, with an approval gate before anything went public.