SR AI Platform Engineer

Position Summary

Assent is building a governed platform that turns AI use cases into production-ready services — usable by technical and non-technical builders alike. You'll build the self-serve builder experience at the heart of that platform: templated build options that let someone pick “a knowledge base agent,” “an internal connector,” or “an Okta-gated web app” and have the platform generate the scaffolding, wire up the deployment pipeline, and enforce security and cost patterns automatically — without them touching Terraform, CI/CD, or a login flow themselves.

This is an infrastructure and platform engineering role with a product mindset: your “customers” are the other builders — engineers and business champions alike — who ship through what you build. You'll spend your time expanding the catalog of self-service build patterns, automating the scaffolding behind each one, and hardening the connector, auth, and cost-tracking primitives every pattern depends on.

Responsibilities include:

• Design, build, and expand the self-serve build patterns builders pick from — a knowledge base agent, an internal or external connector, an Okta-gated web app, a workflow automation — each backed by automated scaffolding rather than a one-off manual build.
• Turn proven, one-off deployment patterns into repeatable, templated build options: automate the repo scaffolding, CI/CD wiring, and infrastructure provisioning behind each one so a builder can go from “I want to build X” to a working deployment without hand-rolling any of it.
• Own and evolve the core deployment pipeline underneath every pattern: containerized apps on AWS ECS Fargate behind an Application Load Balancer, with SSO enforced at the listener so no pattern — and no builder — has to implement their own login.
• Write and maintain Infrastructure-as-Code (Terraform) for compute, networking, and IAM, kept strictly separate from application code so infra changes stay reviewable and auditable on their own.
• Design and harden connector and integration patterns — including MCP-style server patterns — so a builder can request a connection to an internal or vendor system and get one that's already scoped, OAuth-secured, and deny-by-default, rather than building auth from scratch each time.
• Bake mandatory logging, metering, and cost-tracking into every pattern and every service the catalog provisions, so cost visibility and governance are automatic rather than something each builder has to remember.
• Partner with cloud operations on anything account- or network-level (load balancer rules, security groups, IAM roles, DNS, container registries), and keep pattern development moving while that work is in flight.
• Review and harden what other builders ship against these patterns before it reaches production — you're a checkpoint for security and reliability, not just a pattern author.

Qualifications

Your Knowledge, Skills and Abilities

We strongly value your talent, energy and passion. It will also be valuable to Assent if you have the following qualifications

Required

• 5+ years of software engineering experience, with meaningful hands-on time in cloud infrastructure and platform engineering — not just consuming a platform someone else built.
• Strong experience with AWS container services (ECS/Fargate or equivalent), load balancers, and IAM — you've run production services, not just deployed a demo.
• Production experience with Terraform or another mainstream IaC tool, including keeping infrastructure and application code in cleanly separated repositories.
• Experience implementing SSO/OIDC at the infrastructure layer (Okta, Auth0, or equivalent) so individual applications don't own their own auth.
• Strong Python scripting ability — comfortable writing automation, tooling, and operational scripts, and reviewing application code.
• Working knowledge of Docker — image builds, multi-stage builds, and registry management — as the packaging layer underneath ECS deployments.
• Experience with CloudWatch (or equivalent) for logging, metrics, and alerting, since the role owns building metering and cost-tracking hooks into every service.
• Working knowledge of API integration patterns and OAuth scoping — you understand the difference between declaring a security constraint and actually enforcing it at runtime.
• Comfort operating in a fast-moving environment where the pattern library is still being written, not just followed.

Strong plusses

• AWS certification (Solutions Architect, DevOps Engineer, or equivalent) demonstrating depth beyond day-to-day tooling familiarity.
• Familiarity with MCP (Model Context Protocol) or similar agent-to-tool connectivity standards.
• Experience building or operating a cost-metering / FinOps layer for cloud services.
• Experience as a platform or DevOps engineer supporting other engineering teams — you think in terms of paved roads and self-service, not one-off builds.
• Exposure to LLM-powered or agentic systems, even if your own hands-on work has been on the infrastructure side rather than the agent logic.