Software Development Engineer III (AI

Platform / Backend)

About FlyCT (Cleartrip, Flyin.com MENA)

At FlyCT, we are building the next generation of intelligent travel experiences powered by AI.

Our mission is to simplify travel using cutting-edge technologies including Large Language

Models (LLMs), Agentic AI, Retrieval-Augmented Generation (RAG), intelligent workflows, and scalable cloud-native platforms.

As an SDE-III, you will play a key role in designing and building AI-powered backend platforms that serve millions of users while ensuring reliability, scalability, security, and performance.

About the Role

As a Software Development Engineer III, you will own critical business domains from product conception to production deployment. You will independently design scalable backend services while also leading the adoption of modern AI engineering practices.

Beyond traditional backend development, you will design intelligent systems powered by LLMs,

AI agents, vector databases, RAG pipelines, and workflow orchestration frameworks. You will collaborate with Product, Data Science, ML Engineering, and Platform teams to deliver production-grade AI capabilities.

You are expected to make architectural decisions, mentor engineers, drive engineering excellence, and ensure that AI systems are reliable, observable, secure, and cost-efficient.

What You’ll Do

Backend Engineering

● Own end-to-end delivery of one or more business domains.

● Convert ambiguous product requirements into scalable technical solutions.

● Design high-level and low-level architecture independently.

● Build highly scalable microservices and distributed systems.

● Optimize existing backend services for latency, throughput, resiliency, and cost.

● Drive engineering best practices around code quality, testing, documentation, and observability.

● Mentor junior engineers and conduct design/code reviews.

AI Engineering

(LLMs).

retrieval.

● Build production-grade Generative AI applications using Large Language Models

● Design Retrieval-Augmented Generation (RAG) architectures for enterprise knowledge

● Build Agentic AI solutions capable of planning, reasoning, and tool orchestration.

● Develop AI-driven workflows using frameworks such as LangGraph and LangChain.

Integrate external tools, APIs, databases, and enterprise systems into AI agents.

●

Implement context engineering strategies to improve response quality and reduce





●

hallucinations.

● Design prompt engineering frameworks, reusable prompt templates, and evaluation strategies.

● Develop semantic search solutions using embeddings and vector databases.

● Build multi-agent systems for complex business workflows.

● Establish AI evaluation pipelines including latency, hallucination detection, grounding,

and answer quality.

● Optimize AI inference cost, token utilization, and response latency.

Platform & Architecture

● Design scalable AI platform architecture.

● Evaluate architectural trade-offs between:

○ RAG vs Fine-tuning

○ Single Agent vs Multi-Agent

○ LangChain vs LangGraph

○ Structured Outputs vs Free-form Generation

○ Vector Search vs Traditional Search

○ Stateless vs Stateful AI Systems

● Build secure AI services with authentication, authorization, guardrails, and responsible AI principles.

● Ensure production readiness with monitoring, tracing, logging, and observability.

Required Technical Skills

Backend

● Strong expertise in Java and Spring Boot.

● Robust understanding of Data Structures & Algorithms.

● Deep knowledge of Object-Oriented Design and Design Patterns.

● Experience building distributed systems and microservices.

● Strong knowledge of REST APIs and asynchronous architectures.

● Experience with Kafka, RabbitMQ, ActiveMQ, or similar messaging systems.

● Experience with Oracle, MySQL, PostgreSQL, MongoDB, or Redis.

● Experience with Docker, Kubernetes, and container orchestration.

● Experience with AWS or GCP cloud platforms.

● Strong understanding of CI/CD pipelines and DevOps practices.

AI / GenAI

Hands-on experience with:

● Large Language Models (OpenAI, Anthropic, Gemini, Claude, Llama, etc.)

● LangChain

● LangGraph

● Retrieval-Augmented Generation (RAG)

● Vector Databases

● Embedding Models

● Semantic Search

● AI Agents

● Agentic AI

● Multi-Agent Systems

● AI-driven Workflow Automation

● Function Calling / Tool Calling





● MCP (Model Context Protocol) concepts (preferred)

● Prompt Engineering

● Context Engineering

● AI Evaluation Frameworks

● Guardrails and Responsible AI

● AI Observability

● AI Cost Optimization

● Token Management

● Streaming Responses

● Conversation Memory Management

Nice to Have

● Experience with Python for AI development.

● Experience with ML lifecycle tools.

● Knowledge of GraphRAG.

● Knowledge of Knowledge Graphs.

● Experience with Neo4j.

● Experience with Vector Search optimization.

● Experience deploying LLMs on cloud infrastructure.

● Experience with Kubernetes-based AI workloads.

● Experience building AI copilots or conversational assistants.

● Experience with OpenTelemetry, LangSmith, or similar tracing platforms.

What We’re Looking For

● 6–10+ years of software engineering experience.

● Strong ownership mindset.

● Excellent system design and architectural skills.

● Ability to evaluate technical trade-offs and make pragmatic decisions.

● Strong communication and stakeholder management skills.

● Experience mentoring engineers.

● Passion for AI and continuous learning.

Key Competencies

● Distributed Systems

● System Design

● Scalability

● Cloud Architecture

● Microservices

● Generative AI

● Agentic AI

● LangGraph

● LangChain

● RAG

● Vector Databases

● Prompt Engineering

● Context Engineering

● AI Workflow Orchestration

● Architectural Trade-offs

● AI Platform Engineering

● Software Craftsmanship

● Leadership & Mentoring

:::

Additional AI-specific interview expectations (recommended to include in the hiring rubric)

For an SDE-3 AI role, evaluate candidates on:

● Designing end-to-end RAG architectures.

● Building production AI agents using LangGraph.

● Choosing appropriate vector databases and embedding strategies.

● Context window optimization and memory management.

● Prompt engineering and context engineering techniques.

● AI evaluation metrics (groundedness, hallucination rate, latency, cost).

● Architectural trade-offs (RAG vs fine-tuning, single vs multi-agent, workflow orchestration

● Production deployment of LLM applications, including observability, monitoring, security,

choices).

and cost optimization.

● AI governance, guardrails, and responsible AI practices.

📌 Software Development Engineer III (Bengaluru)
🏢 Flyin.com
📍 Bengaluru