Lead Engineer AI

Pine Labs; Dadri; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

Pine Labs is seeking a hands-on LeadAI Engineer to build and operate production-grade LLM applications, RAG pipelines and autonomous agent systems for fintech and payment products. You will own solutions from research through deployment and production support, with a security-first mindset.

You will design and implement end-to-end AI features, integrate APIs like OpenAI or Gemini, develop async Python services, secure multi-tenant APIs and monitor performance with OpenTelemetry, Prometheus and

Qualifications

• Production experience with LLM applications and APIs.
• LangGraph and LangChain.
• RAG: ingestion, chunking, embeddings, vector search, reranking and retrieval.
• Vector databases such as pgvector, Pinecone, Weaviate, Milvus or Qdrant.
• Tool calling, structured outputs, agent state, memory and routing.
• Prompt and context engineering.
• LLM evaluation covering accuracy, groundedness, hallucination, latency and cost.
• AI guardrails, prompt-injection protection and secure tool execution.
• LLM tracing and observability using LangSmith, Langfuse or equivalent.
• Model benchmarking and production optimization.
• Strong Python 3.11+, async/await, FastAPI, Pydantic and httpx.
• PostgreSQL, Redis, SQLAlchemy 2.x and Alembic.
• REST APIs, webhooks and event-driven distributed systems.
• OAuth2, JWT, RBAC, multi-tenancy and secret management.
• pytest, integration testing, contract testing and performance testing.
• Docker, Kubernetes, Helm and GitHub Actions.
• OpenTelemetry, structured logging, Prometheus and Grafana.

Responsibilities

• Build production-grade LLM applications and agents for fintech products.
• Integrate OpenAI, Gemini or similar APIs with tool calling and streaming.
• Implement prompt engineering, model evaluation and guardrails.
• Develop async Python services with FastAPI, Pydantic, SQLAlchemy and Alembic.
• Design secure, multi-tenant APIs using OAuth2, JWT, RBAC with OWASP practices.
• Deploy and operate services using Docker, Kubernetes, Helm and CI/CD.
• Monitor AI performance with OpenTelemetry, Prometheus and Grafana.
• Own features end-to-end and contribute to architecture and code reviews.

Skills

LLM Apps

LangGraph

LangChain

RAG Pipelines

Vector Databases

Tool Calling

Prompt Engineering

Python

PostgreSQL

Redis

SQLAlchemy

REST APIs

OAuth2

RBAC

Docker

Kubernetes

CI/CD

OpenTelemetry

Grafana

Education

Bachelor's degree in Computer Science or related field

Tools

LangSmith

Langfuse

GitHub Actions

Job description

We are looking for a hands-on LeadAI Engineer to build and operate production-grade LLM applications, RAG pipelines and autonomous agent systems for fintech and payment products.

You will own solutions from research and architecture through development, deployment, evaluation and production support. Strong AI expertise, Python engineering and a security-first mindset are essential.

Key Responsibilities

• Build production-grade RAG pipelines and agentic workflows using LangGraph and LangChain.
• Integrate OpenAI, Gemini or similar LLM APIs using tool calling, structured outputs and streaming.
• Implement prompt engineering, model evaluation, guardrails and LLM observability.
• Develop async Python services using FastAPI, Pydantic, SQLAlchemy and Alembic.
• Build secure, multi-tenant APIs using OAuth2, JWT, RBAC and OWASP practices.
• Design reliable event-driven services, payment integrations and agent connectors.
• Deploy and operate services using Docker, Kubernetes, Helm and CI/CD.
• Monitor application and AI performance using OpenTelemetry, Prometheus and Grafana.
• Use AI coding tools such as Cursor, Copilot or Claude Code to improve delivery speed.
• Own features end-to-end and contribute to architecture, code reviews and technical decisions.

Must-Have AI Skills

• Production experience with LLM applications and APIs.
• LangGraph and LangChain.
• RAG: ingestion, chunking, embeddings, vector search, reranking and retrieval.
• Vector databases such as pgvector, Pinecone, Weaviate, Milvus or Qdrant.
• Tool calling, structured outputs, agent state, memory and routing.
• Prompt and context engineering.
• LLM evaluation covering accuracy, groundedness, hallucination, latency and cost.
• AI guardrails, prompt-injection protection and secure tool execution.
• LLM tracing and observability using LangSmith, Langfuse or equivalent.
• Model benchmarking and production optimization.
• Strong Python 3.11+, async/await, FastAPI, Pydantic and httpx.
• PostgreSQL, Redis, SQLAlchemy 2.x and Alembic.
• REST APIs, webhooks and event-driven distributed systems.
• OAuth2, JWT, RBAC, multi-tenancy and secret management.
• pytest, integration testing, contract testing and performance testing.
• Docker, Kubernetes, Helm and GitHub Actions.
• OpenTelemetry, structured logging, Prometheus and Grafana.

Preferred Skills

• Fintech or payment-system experience.
• TypeScript, Node.js, React, Vite and SDK development.
• PCI-DSS, SOC 2 or RBI payment guidelines.
• Kotlin, Ktor or Gradle.

Experience and Mindset

• 5–10 years of software-engineering experience.
• Minimum 2+ years of hands-on GenAI, RAG or agentic AI experience.
• Experience deploying at least one LLM-based application to production.
• Strong ownership, research ability and product thinking.
• Comfortable working independently in a fast-moving environment.
• Security, reliability and auditability focused.