DIA-AI Engineer

PwC; Hyderabad, Bengaluru; Hybrid; INR 2.500.000 - 6.000.000; Vollzeit

Zusammenfassung

PwC is seeking a hands-on AI/ML engineer in Hyderabad to architect and deliver production-grade AI solutions. You will design agents and multi-agent systems, implement end-to-end pipelines, and ensure robust observability and security across the stack.

You will work with PyTorch, TensorFlow, LangChain, Claude, GPT, and cloud platforms to deploy scalable AI capabilities in enterprise environments.

Qualifikationen

• Hands-on experience in ML, DL, Generative AI, and AI agents with production deployments.
• Proficient in model development using PyTorch, TensorFlow, Scikit-learn, and Hugging Face.
• Experience building LLM-driven apps with Claude, GPT, and open-source models.
• Strong background in RAG, embeddings, vector DBs, and semantic search.
• Experience designing AI agents and multi-agent systems with modern frameworks.
• Proficient in agent orchestration, planning, reasoning, tool calling, memory, and human-in-the-loop.
• Hands-on with MCP/A2A protocols, API integrations, and enterprise tool connectivity.
• Deep understanding of AI agent harness architecture, loops, context, guardrails, error recovery.
• Experience building AI evaluation frameworks and observability tools.
• Mandatory full-stack development and cloud engineering experience (Python/JS, AWS/Azure/GCP, Docker, Kubernetes).
• Strong MLOps, DevOps, CI/CD, Terraform, testing, security, and production monitoring.
• Real-world end-to-end production delivery experience.

Kenntnisse

Machine Learning

Deep Learning

Generative AI

Agentic AI

LLM-powered applications

RAG architectures

Embeddings

Vector databases

Context engineering

AI agents

LangGraph

LangChain

CrewAI

AutoGen

agent orchestration

planning

reasoning

tool calling

memory

state management

human-in-the-loop

MCP

A2A protocols

API integrations

enterprise tool connectivity

AI agent harness architecture

execution loops

context management

guardrails

error recovery

AI evaluation frameworks

LLM-as-a-Judge

RAG quality

hallucination detection

agent task success

tool-call accuracy

AI observability tools

LangSmith

Langfuse

Ragas

DeepEval

Arize Phoenix

Full-stack development

Python

Java

TypeScript

REST APIs

microservices

React

Angular

Cloud engineering

AWS

Azure

GCP

Docker

Kubernetes

serverless

CI/CD

Terraform

security

production monitoring

MLOps

DevOps

end-to-end production delivery

Tools

PyTorch

TensorFlow

Scikit-learn

Hugging Face

Claude

GPT

Open-source models

enterprise AI platforms

LangGraph

LangChain

CrewAI

AutoGen

RAG architectures

MCP

A2A protocols

API integrations

LangSmith

Langfuse

Ragas

DeepEval

Arize Phoenix

AWS

Azure

GCP

Docker

Kubernetes

serverless

Terraform

CI/CD

security

production monitoring

Jobbeschreibung

• Strong hands-on experience in Machine Learning, Deep Learning, Generative AI, and Agentic AI, with proven production implementations.
• Experience building, training, fine-tuning, evaluating, and deploying ML models using PyTorch, TensorFlow, Scikit-learn, and Hugging Face.
• Hands-on experience developing LLM-powered applications using Claude, GPT, open-source models, and enterprise AI platforms.
• Strong experience implementing RAG architectures, embeddings, vector databases, semantic search, and context engineering.
• Proven experience designing and building AI agents and multi-agent systems using LangGraph, LangChain, CrewAI, AutoGen, or equivalent frameworks.
• Experience implementing agent orchestration, planning, reasoning, tool calling, memory, state management, and human-in-the-loop workflows.
• Hands-on experience with MCP, A2A protocols, API integrations, and enterprise tool connectivity.
• Strong understanding of AI agent harness architecture, including execution loops, context management, tool permissions, guardrails, and error recovery.
• Experience building AI evaluation frameworks covering LLM-as-a-Judge, RAG quality, hallucination detection, agent task success, and tool-call accuracy.
• Hands-on experience with AI observability and evaluation tools such as LangSmith, Langfuse, Ragas, DeepEval, or Arize Phoenix.
• Mandatory full-stack development experience with Python, Java/TypeScript, REST APIs, microservices, and frontend frameworks such as React or Angular.
• Mandatory cloud engineering experience with AWS, Azure, or GCP, including Docker, Kubernetes, serverless architectures, and cloud-native deployments.
• Strong experience in MLOps, DevOps, CI/CD, Terraform, automated testing, security, and production monitoring for AI applications.
• Demonstrated real-world, end-to-end production delivery experience, including architecture, development, deployment, scalability, performance optimization, and operational support.
• Bonus: Anthropic certifications or training, Claude Code, Claude Agent SDK, custom agent harnesses, agent skills, subagents, hooks, and AI-powered developer tooling.