About the Role The L2 Engineer – Platform & Agentic Ops bridges core infrastructure operations, platform reliability, and autonomous AI agent operations (Agentic Ops). In this intermediate role, you act as the operational backbone for enterprise platforms hosting microservices, data workflows, and multi-agent AI systems. You will handle escalations from Level 1 support, diagnose platform anomalies, trace AI agent execution failures (e.g., tool-call loops, prompt regressions, and token overruns), and maintain high availability across cloud environments.

This position works closely with Platform, DevOps, and AI/ML Engineering teams to automate runbooks, optimize agentic workflows, and maintain platform stability.

Responsibilities

- Agentic Ops & Telemetry Analysis: Monitor autonomous AI agents and execution runtimes. Debug multi-step agent workflows, handle failed tool calls/function calling, resolve context-window and token budget limits, and mitigate runaway loops or infinite retry chains.
- Platform Operations & Incident Management: Serve as an escalation point (L2) for production incidents, service degradations, and system alerts across Kubernetes clusters, API gateways, and cloud infrastructure.
- Observability & Distributed Tracing: Instrument and monitor end-to-end telemetry across both traditional infrastructure and LLM/agent pipelines (inspecting spans from prompt $rightarrow$ routing $rightarrow$ retrieval $rightarrow$ tool execution).
- Automation & Runbook Engineering: Convert repetitive manual fixes into automated scripts, self-healing jobs, or standardized Standard Operating Procedures (SOPs)



to empower L1 teams.
- Cost & Quota Governance: Track, manage, and enforce rate limits, model routing thresholds, and API credit/token consumption across foundation model providers and vector databases.
- CI/CD & Deployment Support: Assist in deploying, validating, and rolling back platform services, prompt configurations, agent definitions, and model gateway updates safely.
- Root Cause Analysis (RCA): Partner with L3/core engineering teams on detailed post-mortems for major platform or agentic pipeline failures.

Requirements
- Experience: 2–4 years in Systems Support, Site Reliability Engineering (SRE), DevOps, or Platform Operations.
- Scripting & Programming: Proficiency in Python (essential for agent runtimes and automations) or Bash/Go/Node.js.
- Cloud & Containerization: Practical experience operating and debugging within containerized environments (Docker, Kubernetes) on major cloud platforms (AWS, Azure, or GCP).
- Agentic & LLM Fundamentals: Foundational understanding of how LLMs and agent frameworks function—including tool/function calling, Retrieval-Augmented Generation (RAG), embeddings, context window management, and prompt handling.
- Observability Stack: Hands-on experience with logging, metric,



and tracing tools (e.g., Datadog, Prometheus, Grafana, OpenTelemetry, ELK).
- Networking & APIs: Solid grasp of REST APIs, gRPC, JSON schemas, webhooks, authentication (OAuth/IAM), and HTTP status debugging.
- Problem-Solving & Communication: Transparent documentation skills, disciplined troubleshooting under SLA constraints, and effective cross-functional communication.

Nice to Have
- Agentic Framework Exposure: Direct operational experience with frameworks like LangChain, LangGraph, CrewAI, AutoGen, or Semantic Kernel.
- LLM Tracing & Evaluation Tools: Familiarity with LLM observability tools such as LangSmith, Phoenix (Arize), Langfuse, or Weights & Biases.
- Vector Databases: Operational familiarity with vector stores (e.g., Pinecone, Qdrant, Milvus, Chroma, pgvector).
- Workflow Orchestrators: Exposure to durable execution engines or workflow tools (e.g., Temporal, Airflow, Prefect).
- Infrastructure as Code (IaC): Basic knowledge of Terraform or Helm for managing platform components.
- Certifications: AWS Certified SysOps Administrator / Solutions Architect Associate, CKA (Certified Kubernetes Administrator), or cloud AI certifications.

Pay: ₹696,000.00 - ₹1,890,000.00 per year

Benefits

- Health insurance
- Life insurance
- Provident Fund

Application Question(s):
- Do you have Experience in Python / Go/ Node.js
- Do you have Experience in Observability & Monitoring?

Experience:
- System Support, SRE: 2 years (Required)

Work Location: Hybrid remote in Coimbatore, Tamil Nadu

📌 Engineer - Platform & Agentic Ops (Coimbatore)
🏢 GUHATEK CONSULTING SERVICES PRIVATE
📍 Coimbatore