EY-GDS Consulting-AIA- Support

L2 Support Engineer – Gen AI Projects (GCP)

Location: Offshore
Role Type: Support & Operations
Experience: 2–4 years

Role Summary

The L2 Support Engineer will monitor, troubleshoot, and support multiple Gen AI applications built on Google Cloud Platform (GCP). The role focuses on following runbooks, identifying where issues are occurring, providing first-level fixes, and handing off deep technical problems to the L3 engineering team.

Key Responsibilities

• Monitor Gen AI pipelines, dashboards, alerts, and system health across various applications.
• Follow runbooks to investigate failures and identify exactly where the process or workflow is breaking.
• Perform basic troubleshooting and resolve issues related to access, permissions, environment setup, or configuration.
• Validate errors using logs, job history, and GCP console (Cloud Functions, Cloud Run, Pub/Sub, Firestore, GCS).
• Raise detailed tickets to L3 with clear findings, logs, and reproduction steps.
• Support daily operations such as checking stuck workflows, failed requests, or incomplete data ingestions.
• Perform minor fixes like restarting jobs, updating config values, clearing queues, or re-triggering pipelines.
• Handle user access requests, IAM role updates, and environment-related incidents within approved guidelines.
• Ensure SLAs are met for incident acknowledgment, analysis, and escalation.
• Maintain documentation, update runbooks, and contribute to process improvements.
• Coordinate closely with L3 engineers, cloud teams, and application owners as needed.

Required Skills

• Basic understanding of cloud environments (preferably GCP).
• Ability to read logs and identify errors across functions, pipelines, and APIs.
• Familiarity with JSON, REST APIs, and debugging simple configuration issues.
• Good understanding of Python basics or the ability to read/debug simple scripts (not mandatory).
• Strong problem-solving skills with an ability to follow structured troubleshooting steps.
• Good communication skills for reporting issues, writing summaries, and escalating to L3.

Nice-to-Have

• Exposure to Gen AI/LLM applications, Document AI, or RAG systems.
• Experience with ticketing systems (Jira, ServiceNow).
• Basic knowledge of Pub/Sub, Cloud Run, Cloud Functions, or Firestore.

L3 Gen AI Engineer – Support & Engineering (GCP)

Location: Offshore
Role Type: L3 Technical Support / Engineering
Experience: 5–10 years (with 2+ years in AI/LLM/RAG-specific systems)

Role Summary

The L3 Gen AI Engineer will provide deep technical support, debugging, and enhancements for enterprise Gen AI applications running on Google Cloud Platform (GCP). This role handles escalations from L2, resolves complex failures across AI pipelines, improves system performance, and implements fixes across models, orchestration, backend services, and integrations.

Key Responsibilities

• Diagnose and resolve complex issues in Gen AI workflows (LLM pipelines, RAG retrieval, document processing, audio-to-text, summarization, etc.).
• Debug model-related problems such as prompt failures, accuracy drops, hallucinations, versioning issues, and latency.
• Troubleshoot backend services including Cloud Run, Cloud Functions, Pub/Sub, Firestore, Composer DAGs, and API integrations.
• Fix failures in document ingestion, Document AI extraction, vector indexing, and data transformation pipelines.
• Review logs, traces, and metrics to identify root causes and apply permanent fixes.
• Implement hotfixes, patch releases, new prompt versions, and workflow improvements.
• Collaborate with UI, middleware, and ML teams to resolve cross-component issues.
• Optimize performance, throughput, token usage, and cloud costs across AI workloads.
• Improve observability by enhancing logging, monitoring, and alerting strategies.
• Manage model lifecycle tasks—deploying new LLM endpoints, tuning prompts, updating embeddings where needed.
• Ensure best practices for security, PII handling, role-based access, and environment integrity.
• Provide guidance to L2 support by updating runbooks, troubleshooting guides, and automation scripts.

Required Skills

• Strong hands-on experience in building or supporting GenAI/LLM-based systems (OpenAI, Vertex AI, Claude, etc.).
• Deep understanding of RAG workflows, vector databases, embeddings, and retrieval pipelines.
• Strong programming skills in Python (must-have).
• Experience with LLM orchestration frameworks (LangChain, LiteLLM, custom orchestrators).
• Good knowledge of GCP cloud services:
• Cloud Run, Cloud Functions
• Pub/Sub, Firestore, GCS
• Vertex AI / Document AI
• Logging, Monitoring, IAM
• Ability to debug distributed workflows, async jobs, and event-driven architectures.
• Experience with CI/CD pipelines and Git-based workflows.
• Strong troubleshooting and root-cause analysis capabilities.

Nice-to-Have

• Experience with multimodal pipelines (Speech-to-Text, audio processing, call summarization).
• Experience deploying or fine-tuning custom models.
• Knowledge of Typescript or middleware frameworks for end-to-end debugging.
• Familiarity with enterprise security, networking, and role-based access controls.