Senior Software Engineer II - AI

Sitetracker; New Delhi; On-site; INR 1,500,000 - 2,100,000; Full time

Job summary

Sitetracker is building Scout, an AI product line that ships production LLM systems and agentic workflows for enterprise customers. As a Product Engineer, you will drive AI capabilities from idea to production across backend and web/mobile experiences, collaborating with frontier models and scalable infrastructure.

You’ll own features end-to-end, work with AWS and Kubernetes, and help deploy AI capabilities into real-world telecom, energy, and infrastructure use cases globally.

Qualifications

• Build production backend services in Python, Java, Node.js or Go.
• Ship production web interfaces with TypeScript and React; React Native is a plus.
• Design REST APIs and data models across relational and NoSQL databases.
• Own features end-to-end: implementation, testing, deployment, and monitoring.

Responsibilities

• Ship customer-facing AI features end-to-end and build agentic LLM systems.
• Work across full stack: backend services and web/mobile frontends.
• Design evaluation pipelines, guardrails, and testing for non-deterministic AI outputs.
• Deploy and operate services on AWS with Docker and Kubernetes.

Skills

Python backend

Java

Node.js

Go

TypeScript

React

React Native

REST APIs

NoSQL databases

Relational databases

AWS

Kubernetes

CI/CD

Tools

LangChain

LangGraph

LlamaIndex

CrewAI

LangSmith

Langfuse

vLLM

GPU inference

Job description

This is more than a full-stack role — it's your chance to build the AI layer of the platform the world's infrastructure runs on. Sitetracker is building Scout, our AI product line: production LLM systems and agentic workflows embedded in the software companies like Cox, Telefnica, and EVgo use to deploy critical infrastructure. As a Product Engineer, you'll take AI capabilities from idea to production agent orchestration and model integration on the backend, and the web and mobile experiences that put them in customers' hands.

This is shipping real, customer-facing AI , not prototypes or research demos. You'll own features end-to-end, work with frontier models (Anthropic, OpenAI), and see your work land in the hands of enterprise users deploying telecom networks, EV chargers, and renewable energy around the world.

What You’ll do:

As a Product Engineer on Scout, you'll hit the ground running and ship customer‑facing AI from day one. You'll design, build, test, deploy, and monitor AI product features end‑to‑end , building agentic LLM systems with multi‑agent workflows, tool/function calling, RAG pipelines, and prompt engineering over frontier models from Anthropic and OpenAI.

You'll work across the full stack: backend services in Python, Java, Node.js, or Go, and TypeScript/React frontends spanning both web and mobile app experiences. You'll make AI dependable by building automated evaluation pipelines, guardrails, and testing strategies for non-deterministic outputs — and running it on production infrastructure: scalable, fault‑tolerant services on AWS and Kubernetes.

You'll think like a product engineer, not a ticket‑taker: dig into user needs, absorb the problem space, make smart calls about what to build, keep stakeholders informed, and find ways to unblock yourself.

The Skills You’ll Have:

Full-Stack Product Engineering

• Build production backend services in one or more of Python (FastAPI or similar), Java, Node.js, or Go.
• Ship production web interfaces in TypeScript and React; mobile app development (React Native or similar) is a plus.
• Design REST APIs and data models across relational and NoSQL databases.
• Own features end-to-end: implementation, testing, deployment, and monitoring.

AI / LLM Engineering

• Build with LLM APIs (Anthropic, OpenAI): agentic workflows, tool/function calling, RAG pipelines, and prompt engineering. Production experience preferred; substantial personal or open-source projects considered.
• Write evaluation pipelines and testing strategies for non-deterministic AI outputs.
• Bonus: agent frameworks (LangChain, LangGraph, LlamaIndex, CrewAI), Model Context Protocol (MCP), LLMOps tooling (LangSmith, Langfuse), vector databases, or model serving (vLLM, GPU inference).

Production Systems & AI Infrastructure

• Deploy and operate services on AWS with Docker, Kubernetes, and CI/CD (Scout runs on AWS).
• Instrument systems with observability and handle customer data securely in AI systems.
• Design fault‑tolerant, scalable services that serve AI workloads reliably.
• Salesforce platform and Apex — Scout integrates with the Sitetracker Salesforce platform.

Product Thinking & Communication

• Dig into user needs and use that context to decide what to build and how.
• Explain technical decisions and AI concepts clearly to technical and non-technical audiences.
• Break challenges into iterative solutions, keep stakeholders informed, and unblock myself.

Within 90 Days, You’ll:

• Ramp up on Scout's architecture, codebase, and AI tooling.
• Ship your first AI-driven product features with support from senior engineers.
• Build a working knowledge of our evaluation pipelines and deployment workflow on AWS.

Within 180 Days, You’ll:

• Independently deliver AI product features end‑to‑end, with increasing complexity.
• Contribute to design discussions and team best practices for secure, scalable AI development.
• Own the quality of your features in production: monitoring, evals, and iteration.

Within 365 Days, You’ll:

• Consistently deliver well‑crafted, end‑to‑end AI product capabilities that increase velocity and throughput.
• Elevate team quality through strong technical work, thoughtful code reviews, and knowledge sharing.
• Serve as a go-to contributor for agentic systems on Scout.