Senior Gen AI Engineer

Role overview

This role sits where GenAI engineering meets learning content. You’ll design and run the agentic systems that generate and quality-check what learners study, and you’ll own the outcomes those systems produce, not just the code behind them. It suits someone who’s as comfortable shipping an agent as they are judging whether the content it created is actually good, and who measures the work by whether learners actually succeed, not just whether it shipped.

What you’ll own

You’ll build the AI systems that generate and quality-check our learning content (lesson scripts, coding exercises, graded projects, MCQs) and own that content through to what reaches learners. The agents are how the work gets done; what you’re accountable for is content that ships quickly, holds up under assessment, and stays current.

• Deliver GenAI content (lesson scripts, coding exercises, graded projects, MCQs) through agentic workflows you build and ship (n8n, LangGraph, LLM APIs), meeting our bar for instructional clarity and technical accuracy and holding an average instructor rating of 4/5 or better.
• Own the quality bar: design the eval sets, rubrics, and guardrails that decide what’s good enough to ship, and author enough content yourself to keep that judgment sharp.
• Keep content accurate and current catch outdated Gen AI concepts, deprecated libraries, wrong syntax, or broken code examples within a week of discovery, and fix them within SLA.
• Contribute to curriculum R&D experiment with new tools and frameworks, and proactively flag what’s worth adding or retiring, with at least one R&D-driven update or recommendation each quarter.
• Build the foundations that keep the pipeline running and improving on its own: memory, knowledge bases, integrations, observability, and clearing the issues that stall generation or loading.

How we’ll measure it:

Content: velocity (more shipped per cycle), efficiency (lower cost and effort per unit), effectiveness (learners who complete it pass what they’re assessed on), and relevancy (it stays current).

Business: skill-assessment success percentage, a minimal delta between content and assessment, and more learners landing in the ideal engagement bands.

Must-have

• 1–2 years building software or AI/automation systems, with at least one deployed, evaluated agentic workflow or LLM-integrated app you can show us (GitHub or live demo, not a tutorial follow-along).
• SME-level working knowledge of GenAI (prompt design, RAG, vector databases, hosted LLM APIs), plus the content ability to judge, produce, and correct technically accurate learning material.
• MERN + Python, enough to build the surfaces your agents plug into and write your own logic and integrations.
• Hands-on with an agent/automation framework (n8n, LangGraph, LangChain, CrewAI, or equivalent), including multi-step, tool-calling workflows.
• An evaluation instinct: you build the rubric before you trust the output, and treat an unmeasured agent as unfinished.
• A feel for where things break, from hallucination and prompt injection to cost at scale and latency, and how to guard against it.
• You use AI tools (Claude, Cursor, ChatGPT, Copilot) as a real part of how you build, every day.
• You write clearly: precise, structured, easy to follow.

Nice-to-have

• EdTech, assessment, or structured-content experience.
• Observability and eval tooling (LangSmith, RAGAS, or similar), and CI/CD for content pipelines.
• A public trail of how you learn: a thoughtful README, a walkthrough, a blog post, an open-source contribution.

Why this role

GenAI content is being figured out as it’s built, and this is a seat where you help shape how it’s done rather than execute someone else’s playbook. You’ll own real systems, see their impact on real learners, and have the room to experiment, ship, and improve fast. If you’ve built something you’re proud of and you’re hungry to build more, we’d love to see it.

Location: Hyderabad (In-office)

Workweek: 5 days

Employment type: Full-time