Platform Architect

Platform Architect · Evergreen Insight Global

HYBRID ON SITE AT INSIGHT GLOBAL HQ IN DUNWOODY M-TH & WFH FRIDAY

We're building an AI delivery platform for the enterprise — reusable agent frameworks, governed multi-tenant SaaS, production-grade AI products. Not one-off builds. Not consulting. Platform.

You will be responsible for:

• Defining what belongs in the core platform vs what gets built custom — and holding that line as delivery pressure mounts
• Designing reusable accelerators, agent frameworks, and integration patterns that get adopted across products and clients
• Architecting multi-tenant enterprise SaaS with governance, security, and observability built in from day one
• Setting engineering standards for agentic systems — how agents are built, deployed, monitored, and governed at scale
• Governing component reuse — classifying every build as platform, template, or custom before engineering starts
• Owning platform reliability — SLAs, observability, incident response, and operational accountability for what you architect
• Partnering with sales and product to translate platform capabilities into repeatable enterprise offers
• Presenting architecture decisions and platform strategy to CIO, CTO, and CDO stakeholders — clearly, without jargon

The experience you bring:

• 10+ years in software architecture and platform engineering
• Built a reusable platform, internal developer platform, or accelerator library that actually got adopted — not just designed
• Multi-tenant SaaS architecture — built from scratch or led a major migration
• Data architecture for AI — vector stores, embedding pipelines, retrieval patterns, tenant data isolation, audit trails, and lineage at scale
• Enterprise integration patterns — API management, event-driven architecture, and connecting AI platforms to complex enterprise stacks (MuleSoft, Apigee, or equivalent)
• SRE mindset — you've owned what you've architected. SLOs, error budgets, incident response, on-call. Platform reliability is your problem too.
• LLMOps — model versioning, evaluation pipelines, prompt management, and inference cost governance across tenants
• Usage metering and tenant cost attribution — how platform consumption gets measured, allocated, and surfaced in a multi-tenant environment
• Canary deployments and model evaluation — safely rolling agent and model updates against production traffic
• Deep cloud-native expertise on Azure, AWS, or GCP
• Experience in regulated industries — financial services, healthcare, or similar
• Shipping track record in Python, TypeScript, Go, or Rust
• Familiarity with AI governance, security, and compliance frameworks — tooling, standards, or hands-on implementation — a plus

If your platform work lives entirely in slide decks and never made it to production, we're probably not a match.

We’re looking for a Platform Architect who designs the core AI platform (the “Factory”) that all customer solutions are built on—not one‑off implementations. This person decides what is reusable vs. bespoke before engineering starts and owns that boundary under real delivery pressure. You’ll define multi‑tenant, cloud‑native architecture for enterprise AI/agentic systems, set standards for APIs, data models, governance, and scaling, and partner closely with delivery teams and sales to ensure everything compounds across clients.

Tech stack: AWS/Azure/GCP, container orchestration (AKS/ACA/ECS), APIs & distributed systems, infrastructure‑as‑code, policy‑as‑code (OPA/Rego), and experience shipping in Python, Go, Rust, or TypeScript—ideally with AI/LLM platforms.