- Build Autonomous Agents: Design and deploy AI agents from scratch to run end-to-end marketing workflows, including content generation, campaign launches, paid channel optimisation, and creative testing.
- Deconstruct Workflows: Break down broad marketing processes into the foundational pieces agents need to succeed: skills, tools, evaluations (evals), guardrails, memory, and feedback loops.
- Develop Evaluation Frameworks: Build robust systems to measure agent quality, catch regressions, and drive autonomous improvement without constant monitoring.
- Design Self-Improving Loops: Ensure agents monitor their own outputs, learn from outcomes, and get sharper over time. The standard: If it works at 3 AM on a Sunday, it should be even better by Monday.
- End-to-End Ownership: Act as both the PM and the Engineer. Own everything from identifying automation opportunities to prototyping, production deployment, monitoring, and iteration.
- Stay at the Frontier: Remain plugged into the cutting edge of agentic AI, new model capabilities, tool-use patterns, multi-agent orchestration, MCP, and eval frameworks—and bring those learnings into production rapidly.
- Create Reusable Infrastructure: Build the internal tooling, infrastructure, and documentation necessary so the rest of the marketing team can trust, operate, and scale what you've built.
AI Agent Operator - Growth Marketing [India]
Job Description
What you'll focus on
What you bring
- Marketing & Growth DNA: You come from a rich growth marketing background and have worn multiple hats across paid ads, email, and organic channels.
- PM Brain, Engineer Hands: You can scope out an MVP, define why it matters, and then actually go build it yourself.
- Living on the AI Frontier: You are deeply immersed in the AI ecosystem. You read the research papers, run the benchmarks, and break the models. You have strong, informed opinions on tool-use vs. code-gen agents, RAG vs. fine-tuning, and how to build evals that actually matter.
- Systems Thinker: You think in compounding loops, not isolated tasks and tickets. You design ecosystems where agents trigger other agents and outputs become inputs.
- Production-Tested: You have built and shipped real agents in production—not just theoretical demos. Your systems handle edge cases, run autonomously, and improve over time.
- Process-Oriented: You know the difference between a cool demo and a reliable agent lies in observability, structured outputs, error handling, and a constant iteration on failure modes.
- Rapid Experimenter: You validate agent architectures with fast, cheap tests before committing to production builds. You thrive on tight feedback loops.
- Autonomous Communicator: You communicate clearly, own your work completely, and never wait to be told what to build next.
- Familiarity with core marketing systems (HubSpot, Google Marketing Platform, or similar GTM tools).
- Deep passion for AI-first marketing tools (e.g., Octave).
- Experience building multi-agent systems with advanced orchestration layers and inter-agent communication.
- A track record of shipping internal tools or platforms that achieved high adoption and reliance from other teams.
What you'll get
- Annual Performance Bonus
- 25 Days of Paid Time Off
- Health Maintenance Organisation (HMO) Coverage from Day One
- Provided MacBook for Enhanced Productivity
- Annual Learning & Development Budget for Professional Growth