Role description

Role Snapshot

PRIMARY FOCUS

AI product engineeringCORE STACK

Python FastAPIFlaskAI EXPERTISE

LLMs RAGDELIVERY

Production ownership

Also looking at P2 with 4 5 Years of Experience

Role Overview

We are seeking a handson Senior AI Software Engineer to design build deploy and operate productiongrade Generative AI capabilities This role is ideal for an engineer who combines strong backend fundamentals with practical experience building solutions spanning Generative AI RAG and Agentic AI

Key Responsibilities

Design and develop Python services that integrate with leading LLM APIs including OpenAI Anthropic or Amazon Bedrock

Build endtoend RAG pipelines covering document ingestion chunking embeddings vector search retrieval prompt construction and response generation

Build agentic AI capabilities incorporating tool execution workflow orchestration memory guardrails and MCPbased integrations

Implement LLMOps practices including observability telemetry performance monitoring and cost optimisation

Develop robust REST APIs and backend services using FastAPI or Flask

Improve response quality by diagnosing prompt behavior retrieval relevance hallucinations context limitations and token constraints

Engineer production resilience through retries ratelimit handling error management observability and operational support

Deploy AIenabled features to cloud environments and take ownership of reliability maintainability and performance

Collaborate with product architecture engineering and quality teams to translate use cases into secure usable and testable solutions

Contribute to code reviews engineering standards technical documentation and continuous improvement of the AI delivery platform

Leverage AI pair programming tools GitHub Copilot Cursor AI Gemini Code Assist to accelerate development code conversion and version upgrades

Required Qualifications





6 years of professional experience in Python backend or software engineering

At least 1 year of handson experience delivering LLMenabled features or applications to production

Experience with agent frameworks such as Microsoft Agent Framework MAF LangChainLangGraph AutoGen or CrewAI

Demonstrated experience building RAG pipelines using embeddings vector databases retrieval strategies and prompt orchestration

Strong proficiency in Python and practical experience with FastAPI or Flask

Production experience integrating LLM APIs such as OpenAI Anthropic or Amazon Bedrock

Practical understanding of prompt debugging contextwindow constraints token management hallucination mitigation and model response evaluation

A track record of shipping userfacing features and supporting them in production

Strong problemsolving debugging communication and crossfunctional collaboration skills

Preferred Qualifications

Experience with LangChain or LlamaIndex

Experience with vector databases such as Pinecone Weaviate or ChromaDB

Experience creating rapid prototypes or demonstrations using Streamlit

Cloud deployment experience on AWS Microsoft Azure or Google Cloud Platform

Familiarity with CICD containerization monitoring security controls and automated testing for AI services

What Success Looks Like

AI features are delivered as working maintainable product capabilities not just prototypes

Services remain reliable under production traffic rate limits failures and changing model behavior





Response quality improves through systematic evaluation retrieval tuning prompt refinement and defect resolution

Code is transparent testable documented and aligned with engineering standards

IDEAL CANDIDATE A handson engineer who enjoys building debugging and operating AI features with a strong bias toward working software and measurable outcomes

Skills

Mandatory Skills : Java Microservices

About LTM
LTM is an AI-centric global technology services company and the Business Creativity partner to the world’s largest and most disruptive enterprises. We bring human insights and intelligent systems together to help clients create greater value at the intersection of technology and domain expertise. Our capabilities span integrated operations, transformation, and business AI — enabling new ways of working, new productivity paradigms, and new roads to value. Together with over 87,000 employees across 40 countries and our global network of partners, LTM — a Larsen & Toubro company — owns business outcomes for our clients, helping them not just outperform the market, but to Outcreate it. Please also note that neither LTM nor any of its authorized recruitment agencies/partners charge any candidate registration fee or any other fees from talent (candidates) towards appearing for an interview or securing employment/internship. Candidates shall be solely responsible for verifying the credentials of any agency/consultant that claims to be working with LTM for recruitment. Please note that anyone who relies on the representations made by fraudulent employment agencies does so at their own risk, and LTM disclaims any liability in case of loss or damage suffered as a consequence of the same. Recruitment Fraud Alert - https://www.ltm.com/careers/recruitment-fraud-alert

📌 Specialist - Software Engineering (India)
🏢 LTM
📍 India