Sr Specialist Software Engineering (AI Forward Deployed Engineer)

AT&T Mobility Services; India; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

AT&T Mobility Services seeks an experienced AI/ML software engineer to architect and deploy GenAI, RAG, and Agentic AI solutions for enterprise use. You will collaborate with stakeholders to convert requirements into production-ready AI systems and integrate them with enterprise tools and cloud services.

The role emphasizes building robust retrieval pipelines, multi-agent orchestration, and secure API-based microservices using Python, FastAPI, and modern AI tooling in a large-scale environment.

Qualifications

• Experience delivering production-grade GenAI solutions.
• Hands-on delivery of AI/ML systems in enterprise environments.
• Strong backend development and API design expertise.

Responsibilities

• Identify high-value AI use cases with stakeholders.
• Design and build Generative AI, RAG, and Agentic AI apps.
• Engineer retrieval pipelines with ingestion, parsing, embeddings and indexing.
• Build single-agent and multi-agent systems with planning, tool calls, and memory.
• Develop MCP servers/clients for enterprise tool integration.
• Create reusable tools, plugins, and connectors for controlled agent execution.
• Develop AI-backed backend services and microservices using Python and FastAPI.
• Integrate AI with enterprise apps, databases, SaaS, and cloud services.
• Apply prompt engineering and context engineering for outputs and memory.
• Create automated evaluation pipelines for retrieval quality and groundedness.
• Implement LLMOps and GenAIOps practices for deployment and monitoring.
• Enforce AI security guardrails to prevent prompt injections and data leakage.
• Rapidly prototype and convert prototypes into scalable enterprise apps.
• Troubleshoot AI, retrieval, data, and integration issues to closure.
• Produce technical designs, runbooks, and deployment docs.
• Collaborate with Product Owners and Engineers to ensure delivery visibility.

Skills

Python programming

Generative AI

LLM applications

RAG patterns

Agentic AI

MCP

Prompt engineering

Embeddings

Vector search

LLMOps

Tools

Azure OpenAI

LangChain

LangGraph

AutoGen

FastAPI

REST APIs

Azure AI Search

PostgreSQL/pgvector

Redis

Qdrant

Job description

Experience

8+ years of relevant software engineering, cloud engineering, AI/ML engineering, platform engineering, or solution engineering experience, including hands‑on delivery of production‑grade Generative AI, RAG, or Agentic AI systems

Key Responsibilities

• Work directly with business and technical stakeholders to identify high-value AI use cases and convert ambiguous requirements into production-ready solutions.
• Design and build Generative AI, Retrieval-Augmented Generation (RAG), Advanced RAG, and Agentic AI applications.
• Engineer retrieval pipelines covering document ingestion, parsing, chunking, metadata enrichment, embeddings, indexing, hybrid retrieval, reranking, grounding, and response generation.
• Build single-agent and multi-agent systems with planning, reasoning, tool calling, workflow orchestration, state management, memory, retries, exception handling, and human-in-the-loop approvals.
• Develop and integrate Model Context Protocol (MCP) servers and clients to securely connect AI agents with enterprise tools, APIs, data sources, and workflows.
• Build reusable tools, functions, plugins, APIs, and connectors for structured and controlled agent execution.
• Develop AI-enabled backend services and microservices using Python, FastAPI, Flask, REST APIs, asynchronous processing, and event-driven patterns.
• Integrate AI solutions with enterprise applications, databases, SaaS platforms, cloud services, security platforms, and knowledge repositories.
• Apply prompt engineering and context engineering, including system prompts, few-shot prompting, structured outputs, context compression, and conversation-memory strategies.
• Build automated evaluation pipelines to measure retrieval quality, groundedness, faithfulness, answer relevance, hallucination, task completion, latency, security, token consumption, and cost.
• Implement LLMOps and GenAIOps practices covering model and prompt versioning, testing, deployment, tracing, observability, rollback, and production monitoring.
• Implement AI security guardrails against prompt injection, indirect prompt injection, jailbreaks, sensitive-data disclosure, insecure tool execution, privilege escalation, and excessive agency.
• Rapidly develop proofs of concept and convert validated prototypes into secure, scalable, resilient, maintainable, and supportable enterprise applications.
• Troubleshoot complex AI, retrieval, application, data, API, cloud, and integration issues and drive them through root-cause analysis and closure.
• Create technical designs, architecture diagrams, evaluation reports, runbooks, deployment documentation, and operational support procedures.
• Collaborate with Product Owners, Architects, Software Engineers, DevOps, Cybersecurity, Data Engineering, and business teams while maintaining clear delivery visibility.

Required Technical Skills

Must Have

• Strong programming experience in Python, including production‑grade backend and AI application development.
• Hands‑on experience building Generative AI and Large Language Model (LLM) applications using Azure OpenAI, OpenAI APIs, or equivalent enterprise model platforms.
• Strong experience with RAG and Advanced RAG patterns, including query rewriting, multi‑query retrieval, metadata filtering, hybrid search, reranking, contextual retrieval, citations, and grounded generation.
• Hands‑on experience with Agentic AI, AI agents, tool/function calling, structured outputs, agent orchestration, and multi‑agent workflows.
• Understanding of Model Context Protocol (MCP), including MCP hosts, clients, servers, tools, resources, prompts, authentication, authorization, and secure enterprise integration.
• Experience with prompt engineering and context engineering, including system instructions, few‑shot examples, context‑window management, memory, and output schemas.
• Strong understanding of embeddings, semantic search, vector search, similarity search, hybrid retrieval, reranking models, and retrieval‑quality optimization.
• Experience with vector stores and search platforms such as Azure AI Search, PostgreSQL/pgvector, Redis, Pinecone, Qdrant, Weaviate, or equivalent.
• Experience with document ingestion and knowledge pipelines, including PDF/Office document parsing, OCR, chunking, metadata extraction, indexing, and access‑control‑aware retrieval.
• Experience with AI frameworks such as Microsoft Semantic Kernel, LangChain, LangGraph, AutoGen, or equivalent orchestration technologies.
• Experience building APIs and microservices using FastAPI, Flask, REST, JSON, asynchronous programming, background workers, and message queues.
• Experience with Microsoft Azure services and cloud‑native application architecture.
• Experience with AI evaluation and Evals, including groundedness, faithfulness, retrieval precision/recall, answer relevance, hallucination detection, regression testing, and LLM-as-a-Judge approaches.
• Experience with LLMOps/GenAIOps, including model configuration, prompt versioning, experiment tracking, tracin