AI Operations Engineer

Chimera Technologies Pte Ltd; Bengaluru; Ibrido; INR 1.800.000 - 2.400.000; Tempo pieno

Vantaggi offerti da questo lavoro

Insurance benefits for self and spouse

Maternity benefits

Innovation based Continuous improv. pl

Opportunity to work with multiple ISVs

Chimera Talks

Fortnightly strategy sessions

Descrizione del lavoro

Chimera Technologies Pte Ltd. seeks an experienced engineer to own production incidents on an AI agents platform end‑to‑end, combining on‑call production support with hands‑on engineering.

You will drive automation, observability, and platform improvements through the client SDLC, while maintaining incident response ownership. The role blends monitoring, troubleshooting, and builds—shifted towards development as automation lands—within a hybrid environment requiring strong coding, observability,

Competenze

• 4+ years supporting production systems in on‑call environments.
• Strong programming in Python and Go with ability to read and debug unfamiliar code.
• Hands‑on with observability stacks (Prometheus, Grafana, Loki) and reading traces.
• Working knowledge of LLM and RAG based applications for agent and chain failures.
• IAM fundamentals: SSO, OAuth, RBAC; scalable access provisioning.
• AWS in production: EKS, EC2, IAM, S3, CloudWatch, VPC basics.
• ArgoCD or similar GitOps tooling; reading sync status and performing rollbacks.
• Linux administration and container fundamentals.
• REST APIs; debugging cross‑service integrations.
• Ticketing/ITSM discipline (ServiceNow or equivalent).
• Willingness to work rotating shifts including nights; potential for 24x7 with weekend cover.
• Clear written communication with diagnostic notes.

Mansioni

• Monitor platform health and act on alerts before users report them.
• Triage incidents and service requests, validate priority, adhere to SLAs.
• Diagnose agent and chain failures—tool calls, retrieval, timeouts, retries, orchestration.
• Troubleshoot API, integration and gateway connectivity issues.
• Provision and troubleshoot access and entitlements—SSO, OAuth, RBAC.
• Investigate issues across logs, metrics and traces.
• Verify deployments and perform rollbacks via GitOps pipelines.
• Apply fixes and known error resolutions; escalate with diagnosed root cause.
• Inspect workflow executions and run histories; document resolutions.
• Build automation to deflect repeats and improve resilience.
• Deliver small config changes and fixes via change/release processes.
• Maintain runbooks, SOPs, and known error databases.

Conoscenze

Python

Go

Prometheus

Grafana

Loki

AWS

ArgoCD

GitOps

Linux

REST APIs

ITSM (ServiceNow)

SSO

OAuth

RBAC

Incident ownership

On-call

Shift work

Strumenti

Prometheus

Grafana

Loki

ArgoCD

Descrizione del lavoro

Shift: 24x5 rotational across three shifts (IST), including a night shift. The engagement runs 24x5 today and is expected to extend to 24x7; candidates must be willing to rotate into weekend cover when it does.

Role type: Hybrid engineering and support. Roughly two-thirds production support and incident ownership, one-third build work — automation, tooling, observability and platform improvements shipped through the client SDLC. The balance shifts further toward build as runbooks mature and automation lands. This is not a pure support role, and not a pure development role.

Role summary

Own production incidents on an AI agents platform end to end — detection, triage, diagnosis and resolution — and build the automation and tooling that reduces recurring work. This is a mixed role: production support and hands‑on engineering in the same seat. Alongside the queue , you will build automation, improveobservability,and ship fixes into the platform. It is not a ticket‑routing role, and it is not support‑only.

Responsibilities

• Monitor platform health and act on alerts before users report them
• Triage incidents and service requests, validate priority, and hold response and resolution SLAs
• Diagnose agent and chain failures — tool‑calling errors, retrieval failures, timeouts, retries, orchestration faults
• Troubleshoot API, integration and gateway connectivity issues
• Provision and troubleshoot access and entitlements — SSO, OAuth, client‑ID allowlisting, RBAC
• Investigate application and infrastructure issues across logs, metrics and traces
• Verify deployments and execute rollbacks through the GitOps pipeline; investigate failed and out‑of‑sync releases
• Apply configuration fixes and known error resolutions; restore service and **escalate with a diagnosed root cause**
• Inspect workflow executions — failed and stuck runs, execution history, replay
• Build automation to deflect repeatable incidents and requests
• Deliver small enhancements, configuration changes and fixes through the client’s change and release process, alongside queue work
• Maintain runbooks, SOPs and the known error database from every resolution

Requirements

• 4+ years supporting production systems in an on‑call environment, with direct incident ownership
• Strong programming ability in Python and Go — reads unfamiliar code and debugs it
• Hands‑on with an observability stack — Prometheus, Grafana and Loki, or equivalent — writing queries and reading traces
• Working knowledge of LLM and RAG based applications, sufficient to diagnose agent and chain‑level failures
• IAM fundamentals: SSO, OAuth, RBAC, access provisioning at scale
• AWS in a production context — EKS, EC2, IAM, S3, CloudWatch, VPC and networking basics
• ArgoCD or equivalent GitOps tooling — reading sync status, verifying releases, executing rollbacks
• Linux administration and container fundamentals
• REST APIs, and debugging integration failures across services
• Ticketing and ITSM discipline — ServiceNow or equivalent
• Willingness to work a rotating shift including nights, and to move to a 24x7 rotation with weekend cover when the engagement extends
• Clear written communication: escalation notes that state what was checked, what was found and what was ruled out

Good to have

• LangChain or a similar agent framework
• Both Python and Go
• Temporal or another workflow orchestration engine
• MCP (Model Context Protocol) and toolkit integration
• Vector databases and embedding pipelines
• Participate in several organization wide programs, including a compulsory innovation based Continuous improvement program providing you platforms to showcase your talent.
• Insurance benefits for the self and the spouse including maternity benefits.
• Ability to work on many products as the organization is focusedon working with several ISVs.
• Fortnightly sessions to understand the directions of each function and an opportunity to interact with the entire hierarchy of the organization.
• Celebrations are a common place - physical or virtual. Participate in several games with your coworkers.
• Voice your opinions on topics other than your work - Chimera Talks.