Platform Engineer — Kentron AI

___________________________________

Equity-based compensation | Early-stage

___________________________________

Kentron is building the trust layer for autonomous AI agents: a platform that converts probabilistic agent behavior into deterministic, auditable outcomes via an immutable, cryptographically chained event log. Every action an agent takes produces tamper-evident proof, enabling audit trails, deterministic replay, and crash recovery across 1,000+ SaaS integrations. We run on TypeScript, Bun, Effect, PostgreSQL 17, Redis, Qdrant, Rocicorp Zero, and AWS via SST — a contemporary stack built for durable, event-sourced execution at scale.

We are looking for a Platform Engineer to help us scale the infrastructure. You will own the systems that keep OpenSandbox isolated and performant, Factory's DAG-based agent orchestration reliable under load, and the receipt chain append-only and tamper-evident as we grow from early customers to enterprise deployments. You bring at least 3 years of hands-on experience building, deploying, or operating agentic systems in production — you have felt what breaks when an AI agent fails mid-task and nobody can reconstruct what happened.





This is a high-ownership, high-trust role at the ground floor of a company solving a genuinely hard problem — and compensation is equity-based, meaning you build alongside us and share in what we create.

What you bring:

- 3+ years working with agentic or autonomous AI systems in production environments
- Strong backend engineering in TypeScript and/or Bun; comfortable with functional effect systems (Effect.ts a plus)
- Experience with event-sourced or CQRS architectures; you understand why append-only logs matter
- Distributed systems fluency: PostgreSQL, Redis, message queues, WebSocket-based real-time sync
- Cloud infrastructure experience on AWS; familiarity with infrastructure-as-code (SST, CDK, or equivalent)
- Comfort with containerized workloads (Docker, ECS) and isolated compute environments
- Experience integrating with third-party SaaS APIs at scale; OAuth and credential brokering patterns
- Vector database exposure (Qdrant or similar) is a plus
- Security-minded: you think about multi-tenant isolation, credential boundaries, and tamper-evidence by default

📌 Platform Engineer (India)
🏢 Kentron AI
📍 India