Senior Software Engineer - Deployment Platform

PLT Engineering; Selangor; On-site; MYR 150,000 - 240,000; Full time

Job summary

PLT Engineering in Malaysia is seeking a platform engineer to design and operate the deployment platform used by hundreds of teams. You will work on multi-region deployments, ML model serving, and safe progressive delivery as the default, not the exception.

Reporting to the Senior Software Engineering Manager, this on-site role in Petaling Jaya requires guiding design, code reviews, RFCs, and mentoring to raise engineering standards.

Qualifications

• Experience delivering platform-scale deployment solutions.
• Strong knowledge of Kubernetes, Istio, and progressive delivery patterns.
• Proven ability to diagnose and fix cross-system production incidents.

Responsibilities

• Manage components of the deployment platform end to end from RFC through design, implementation, rollout, and operation.
• Extend the deployment model and platform to support multi-deployment placement, working with OCM for placement and hub-side Istio orchestration for endpoint aggregation.
• Build and harden progressive-delivery mechanics including canary and blue/green rollouts and the guardrails for safe rollout.
• Design deployment abstractions that hold up across heterogeneous workloads, including ML model serving.
• Debug hard cross-system production issues across deployment, autoscaling (Karpenter/KEDA), and service mesh (Istio).
• Raise the engineering bar through design reviews, code review, and RFCs; mentor engineers and help set technical direction.

Skills

Distributed systems

Platform-scale engineering

RFC/design reviews

Mentoring

Tools

Istio

Flagger

Karpenter

KEDA

Blue/Green rollout

Canary

ML model serving

EKS

Job description

Get to Know the Team

We build and operate the deployment platform that ships over thousands of services onto Grab's Managed EKS (MEKS) fleet. Our systems — Conveyor and Flagger-driven progressive delivery — are the path every service change takes to production. In H2, we're extending that path from single-cluster into multi-cluster and multi-region deployments. We're converging ML model serving onto MEKS, and building the deployment control plane. This control plane makes safe, progressive rollouts the default rather than the exception.

Get to Know the Role

This is a platform role at scale: the abstractions you design are used by hundreds of engineering teams, and the failure modes you prevent are the ones that show up as customer-facing incidents. You will be reporting to Senior Software Engineering Manager

This role is onsite based in our Petaling Jaya, Malaysia office.

The Critical Tasks You Will Perform:

• You will manage components of the deployment platform end to end — from RFC through design, implementation, rollout, and operation. You'll take a problem like multi traffic orchestration or canary safety and guide it to landing, not just to merge.
• You will extend the deployment model and platform to support multi-deployment placement, working with OCM for placement and hub-side Istio orchestration for endpoint aggregation.
• You will build and harden progressive-delivery mechanics. These mechanics include Flagger-based canary and blue/green rollouts, and the guardrails that make them safe under real-world conditions. The guardrails ensure rollback correctness, autoscaler interactions, and traffic-shift integrity.
• You will design deployment abstractions that hold up across heterogeneous workloads, including standardising ML model serving workloads onto the same platform.
• You will debug hard, cross-system production issues. These issues involve the interaction of deployment, autoscaling (Karpenter/KEDA), and the service mesh (Istio). You will turn the findings into durable platform fixes rather than one-off patches.
• You will raise the engineering bar through design reviews, code review, and RFCs; mentor engineers and help set technical direction within the team.