Founding AI Engineer

Zstate AI; Kolkata District; On-site; INR 2,500,000 - 7,000,000; Full time

Benefits offered by this job

Competitive compensation

Direct impact from day one

Job summary

Zstate AI is seeking a founding AI Engineer in Gurugram to turn enterprise workflows into production-grade agents. You will design multistep AI agents, use tools, retrieve context, and manage state from prototype to deployment.

You will work directly with the founders, shaping both the product and the company. The role requires hands-on AI engineering, strong Python/TypeScript skills, and end-to-end ownership from day one.

Qualifications

• 3–8 years of experience in backend or fullstack shipping production systems.
• At least 1 year hands-on AI engineering: building agents, LLM workflows.
• Proficiency in Python and/or TypeScript; cloud experience preferred.
• Experience with LangChain and LangGraph and tool integration.
• Strong prompt engineering and evaluation approach.

Responsibilities

• Design and build multistep agents that use tools, retrieve context, hold state.
• Shape product from prototype to production deployment the client relies on.
• Collaborate with founders and enterprise clients on technical scope.
• Implement observability, tracing, logging, and error analysis.
• Translate client workflows into technical specifications for delivery.

Skills

Backend Engineer

AI Engineering

Python & TypeScript

End-to-end ownership

Tools

LangChain

LangGraph

APIs

Job description

Founding AI Engineer

Location: Gurugram, India

Full-time

About Zstate

As AI models become increasingly capable, the industry needs better ways to measure reasoning, reliability, and real-world performance.

Zstate AI is building the evaluation and posttraining infrastructure powering the next generation of AI systems. We create expertdriven benchmarks, reward signals, RL environments, and highquality datasets that help frontier AI models become more capable, reliable, and trustworthy across diverse realworld domains.

Zstate AI is led by a founding team from IITs, IIMs, and DCE, bringing together deep technical and strategic expertise to build categorydefining AI infrastructure. We are backed by Giraffe Studios, a venture studio focused on deeptech startups founded by Himanshu Aggarwal (Cofounder & CEO, Aspiring Minds) and Mohit Tandon (Cofounder, Delhivery).

As one of our earliest hires, you'll work directly with the founders to solve challenging engineering problems and help shape both the product and the company from day one.

What this role looks like

You are one of the first engineers at Zstate AI. Our enterprise clients hand us real workflows that take too much expert time, and your job is to turn those workflows into agents that work. You will design and build multistep agents that use tools, retrieve context, hold state, and complete the work end to end, from an early prototype to a production deployment the client actually relies on.

A day in the life

• Spend your mornings sketching agent architectures on a whiteboard or in a notebook: subagents, routing, handoffs, planning, and state management for long tasks.
• Build those agents with LLMs, tool use, retrieval, and memory in LangChain, LangGraph, or whatever framework fits the problem best.
• Write evaluation frameworks that let you iterate on prompts and architecture using LLMasjudge and deterministic evaluators, so you can prove the agent is improving.
• Integrate the agent with the client systems it needs to talk to: APIs, databases, vector stores, MCP, and other tool integrations.
• Ship it to production, then set up observability, tracing, logging, and error analysis so you can watch it behave in the real world.
• Sit with clients and domain experts, translate their workflow into a spec, and keep iterating as production feedback comes in.

Who you are

• 3-8 years of experience in backend or fullstack engineering, with a real track record of shipping production systems.
• At least one year of handson AI engineering: building agents, LLM workflows, or similar.
• Practical experience with LangChain, LangGraph, or similar frameworks, including tool use and multistep reasoning.
• A habit of using prompt engineering and evaluation frameworks to iterate on behaviour with real feedback.
• Strong proficiency in Python and/or TypeScript, plus APIs, system design, and preferably cloud experience.
• The instinct to own a project end to end, from architecture to a client using it in production.
• High ownership, strong execution, and the tendency to improve systems without waiting to be asked.

Extra credit

• Experience with RAG patterns, vector stores, and knowledge retrieval.
• Experience with agent observability tooling like LangSmith and debugging agents in production.
• Experience working directly with enterprise clients on technical scoping and delivery.
• Exposure to RLHF, SFT, or posttraining workflows.

Why this matters

• Competitive compensation for the right candidate.
• You will ship agentic systems that real enterprise clients use in production, not features that sit on a roadmap.
• Early access to the techniques and frameworks behind the most advanced agentic systems in the world.
• High ownership from day one: no ticket queues, no layers, direct impact.
• A culture that rewards engineering and research depth and curiosity over credentials.