Forward Deployment Engineer (DevOps, AI Deployment)

PwC Acceleration Centers; Hyderabad; Hybrid; INR 4,000,000 - 6,000,000; Full time

Job summary

PwC Acceleration Centers is seeking a senior DevOps Engineer to own how AI solutions are deployed into a client's environment in Hyderabad. You will design pipelines and infrastructure, harden AI applications for production, and ensure enterprise security and governance on AWS.

You will lead deployment architecture, CI/CD, IaC, and release standards across engagements, while guiding integration into regulated environments and building scalable model serving with robust observability and cost

Qualifications

• Ownership of production deployments for AI/LLM deployments.

Responsibilities

• Own the deployment architecture for AI solutions on AWS.
• Design and own CI/CD, Infrastructure as Code, and release standards across engagements.
• Lead integration of AI solutions into legacy and regulated environments, respecting identity, security, and governance.
• Set up scalable model and agent serving, with the vector and retrieval infrastructure behind it.
• Establish observability, evaluation, and cost controls for AI workloads in production.
• Define a practical approach to security, governance, and responsible AI for deployments.
• Build reusable deployment accelerators, and mentor engineers.
• Bring field learnings and product gaps back to the wider practice.

Skills

CI/CD

Docker

Kubernetes

Terraform

IaC

Security IAM

LLMOps

Python

Go

Observability

Tools

AWS

GitHub Actions

GitLab CI

Jenkins

ArgoCD

Prometheus

Grafana

OpenTelemetry

Langfuse

Job description

Experience Required

6 – 9 years.

Job Summary

A senior DevOps engineer who owns how AI solutions are deployed into a client's environment. As the technical owner for deployment, you will design pipelines and infrastructure, harden AI applications for production, and meet enterprise security and governance requirements on AWS.

Key Responsibilities

• Own the deployment architecture for AI solutions on AWS.
• Design and own CI/CD, Infrastructure as Code, and release standards across engagements.
• Lead integration of AI solutions into legacy and regulated environments, respecting identity, security, and governance.
• Set up scalable model and agent serving, with the vector and retrieval infrastructure behind it.
• Establish observability, evaluation, and cost controls for AI workloads in production.
• Define a practical approach to security, governance, and responsible AI for deployments.
• Build reusable deployment accelerators, and mentor engineers.
• Bring field learnings and product gaps back to the wider practice.

Required Qualifications

• Substantial DevOps or platform engineering experience with ownership of production deployments.
• Deep CI/CD, Docker, and Kubernetes experience, with strong Terraform / IaC.
• Strong grounding in identity, security, and networking, and enterprise integration.
• Solid automation skills and a habit of codifying build and run processes.
• Deep, hands-on experience deploying LLM and agentic applications to production (LLMOps), including serving, scaling, retrieval infrastructure, observability, evaluation, and responsible AI.

Preferred Qualifications

• Enterprise AI platforms (Palantir Foundry, Databricks, Snowflake) and MLOps tooling at scale.
• Experience in regulated industries.
• SRE or reliability experience.
• Prior consulting, customer success, or forward-deployed work.
• AWS Certified DevOps Engineer – Professional and/or AWS Certified Solutions Architect – Professional; CKA or a cloud AI/ML certification.

Technical Skills & Tools

• Cloud (AWS): Bedrock, SageMaker, Lambda, ECS, EKS, Step Functions, S3, API Gateway, IAM, CloudWatch
• CI/CD: GitHub Actions, GitLab CI, Jenkins, ArgoCD (GitOps)
• AI deployment (LLMOps): model and agent serving and scaling, RAG & vector databases, evaluation, prompt versioning
• Observability & cost: OpenTelemetry, Langfuse, Prometheus, Grafana
• Security & governance: IAM, secrets management, network security, responsible-AI controls
• Scripting: Python, Go, Bash
• Good to have: MLOps at scale (MLflow, model registries, feature stores), Databricks, Snowflake, Palantir Foundry
• Takes ownership of deployment outcomes.
• Clear communication with client stakeholders.
• Mentors engineers and sets standards.
• Sound judgement on security, governance, and responsible AI.