Skip to job details
pathw. Explore opportunities ↗
Menu

YOUR NEXT OPPORTUNITY

AI Engineer

Pocketpills

Gurgaon, Haryana, India India Source: OnJob

Job description

Description supplied through OnJob, with formatting cleaned for readability.

We're seeking an experienced AI Engineer to design and deploy production-grade LLM applications and AI agents. You'll transform complex AI solutions from prototype through production, building scalable agentic workflows and intelligent systems that drive real-world impact.
Key Responsibilities
Design and implement intelligent agents and workflows using frameworks like LangChain, LangGraph, or similar technologies

Engineer sophisticated multi-step workflows incorporating tool calling, structured outputs, state management, memory systems, and dynamic loops

Develop robust RAG systems including embeddings, vector search, retrieval optimization, and intelligent reranking

Build comprehensive evaluation frameworks to assess accuracy, reliability, latency, and cost efficiency

Establish production-ready observability, guardrails, logging, and error handling mechanisms

Own the end-to-end journey of AI solutions from initial prototype to full-scale production deployment

Required Qualifications
Python proficiency is essential

Strong software engineering fundamentals: OOP principles, asynchronous programming, REST APIs, and robust error handling

Proven hands-on experience with LLM and AI agent development

Expertise in prompt engineering, loop design, and agentic graph architecture

Advanced knowledge of tool calling, function calling, structured outputs, and context/memory management

Deep understanding of RAG systems, embeddings, vector search, and retrieval techniques

Production experience with LangChain, LangGraph, LlamaIndex, or equivalent frameworks

Proficiency with REST APIs and backend service architecture

Experience with vector databases (pgvector, Pinecone, Weaviate, Qdrant, or similar)

Hands-on Docker and cloud infrastructure expertise

Preferred Qualifications
Experience with LangSmith or equivalent observability platforms

Knowledge of Model Context Protocol (MCP)

Expertise in agent evaluation, benchmarking, and automated testing

Familiarity with event-driven and asynchronous architectural patterns

Experience
3–6+ years of software engineering experience with hands-on LLM and Generative AI development.

Source: View this vacancy at the source.

PATHW CAREER INSIGHTS

Source reviewed 27 October 2026

Career insights: AI Engineer

Pathw’s AI-assisted application guidance, reviewed before publication and separate from the employer description. Suggestions do not add employer requirements or confirm current hiring.

What this role involves

This role appears to focus on building production-grade AI applications and agent workflows end to end, from prototype to deployment. You could frame your experience around designing multi-step LLM systems, improving reliability, and shipping scalable backend services that support real-world use.

A strong application would likely show how you have turned experimental AI ideas into robust products, especially where you have handled orchestration, state, memory, observability, and error handling in production.

Sources: Original vacancy

Skills and experience to demonstrate

You may want to emphasize Python, solid software engineering fundamentals, and hands-on work with LLM agents or generative AI systems. Relevant strengths include prompt design, tool/function calling, structured outputs, loop and graph-based agent design, and RAG pipelines with retrieval tuning.

It would also help to highlight practical experience with REST APIs, asynchronous programming, vector databases, Docker, cloud deployment, and evaluation or monitoring approaches that improve latency, accuracy, and cost efficiency. If you have used LangChain, LangGraph, LlamaIndex, LangSmith, or MCP, mention them as examples rather than as strict necessities.

Sources: Original vacancy

Preparing your application

For your resume and interview prep, try to describe AI projects using a clear before-and-after structure: problem, approach, tools, production constraints, and measurable outcomes. Focus on what you built, how you tested it, and how you made it dependable at scale.

You could prepare a few concise stories about debugging agent failures, improving retrieval quality, adding guardrails, or reducing inference cost. If you have only prototype experience, show how you approached production readiness through logging, monitoring, testing, and incremental rollout.

Sources: Original vacancy

Questions for the recruiter

You may want to ask how the team evaluates agent quality in production, what observability and guardrail tools are already in place, and how they decide when an AI workflow is ready to move from prototype to deployment.

Other useful questions could cover the main vector store or orchestration stack, how they handle memory and tool calling across workflows, and whether the role leans more toward platform engineering, experimentation, or product delivery.

Sources: Original vacancy

Check current availability, contract details and applicant eligibility with the original recruiter or employer.