Founding Platform Engineering Lead

What you'll own

• The platform that delivers in 4 weeks. Most companies in our space take 12 months to deliver an implementation. The platform you own compresses three things in parallel: the data layer against the customer's source systems, the agent configuration against their workflows, and the security and sync infrastructure each customer inherits. Today, a complex deployment can take us 10-12 weeks to get done, down from 12 months. Your job is to take it lower, and keep it there as the customer base grows from 10 to 40 to 400.
• The deployment automation. A pipeline of agents that does the work an engineering team would otherwise do manually for every customer. You own how they're designed, how they're evaluated, where humans stay in the loop, and how the next leap in deployment speed comes out of the same team.
• The security platform. Compliance, RBAC, encryption, audit trails, agent guardrails. You design the primitives once and own the security posture as we scale into the largest education systems in the country.
• The connector and data engineering strategy. Every customer brings a new source system, a new schema, a new data contract. You set the strategy for how Risely meets that complexity, and build the framework every new connector inherits.

Who you are

• Systems thinker. You design at the level of the whole platform. You see dependencies before you change anything.
• Technically strategic. You solve hard business problems with platforms and systems. When a customer brings a new compliance requirement, a new data shape, or a new integration target, you design the abstraction that makes the next ten customers with the same problem ship on the platform you already built.
• Abstraction taste. You know which patterns deserve to become primitives. You know when to invest in the platform layer and when to leave a customer feature as bespoke code.
• Investment judgment. You know when to stop adding surface area and reinforce the platform underneath. You've made that call and you've been right.
• Master of many threads. You run several concurrent workstreams without dropping the ball on any of them and you’re able to keep the team posted on all threads at any time
• AI-native. You orchestrate Claude Code agents to multiply yourself. You ship more in a week as a one-person team than most companies ship in a month with five. As we grow into a small squad, you teach the same way of working.

Experience

• Backend. Python and TypeScript in production. Postgres, schema design at scale, query performance under real load.
• Infrastructure. AWS or GCP, Kubernetes or equivalent, CDK preferred. You understand how to make the calls on cost, latency, blast radius, and failure modes.
• Data engineering. Large-scale data pipelines, ETL or ELT, real-time sync, schema evolution. You've integrated against third-party source systems (Salesforce, Workday, Banner, Snowflake, or comparable) and lived with the edge cases.
• Multi-tenant patterns. Tenant isolation, encryption at rest and in transit, RBAC, audit trails. You have a position on shared versus isolated infrastructure and per-customer cost accounting.
• LLMs and agents. You've built with leading foundation model APis, designed evaluation pipelines for non-deterministic systems, and shipped agent products to production. You have a working point of view on what an agent should own and where humans need to stay in the loop.

How to apply