You'll own the platform our agents run on — event-driven architecture, data pipelines, integrations with customer accounting systems, and the infrastructure that makes everything reliable, auditable, and fast.
What you'll do
Event store — Append-only PostgreSQL with tamper-evident hash chaining, real-time event distribution, agent replay and recovery, schema versioning.
Cryptographic signature infrastructure — Key management, signature verification on all agent events, approval chains for external actions.
Workflow Constraint Engine — The gateway that verifies authorization before any external action (email, payment, API call) proceeds.
PII Vault & GDPR compliance — Isolated, encrypted storage for personal data. Right to erasure. Data residency per region.
Multi-tenancy — Row-level security, tenant isolation across all data stores, tenant-scoped LLM API keys.
Authorization (Cerbos) — Attribute-based access control for agents and humans. Deterministic policy evaluation — LLMs never make auth decisions.
MCP security layer — Authentication, capability-based scoping per partner, and rate limiting for the Model Context Protocol endpoint.
Database architecture — PostgreSQL schema design, indexing strategy, performance tuning, read replica topology.
What we're looking for
3+ years experience with Python and/or TypeScript in production.
5+ years designing backend platforms for production systems.
Strong systems design skills — this is the primary requirement. You think in boundaries, contracts, failure modes, and scaling characteristics.
Deep PostgreSQL knowledge — RLS, indexing tradeoffs, event sourcing patterns, schema evolution.
Strong application security architecture skills: encryption models, key management, auth patterns.
Experience with policy-as-code authorization (Cerbos, OPA, Cedar, or similar).
Strong code review skills for security-sensitive AI-generated code.
Nice to have:
Cryptographic primitives in production (digital signatures, hash chaining, application-layer encryption).
Event streaming platforms (Kurrent, Redpanda, Kafka).
SOC 2 or GDPR compliance engineering.
Fintech or regulated-environment background.
Skills Required
- 3+ years experience with Python and/or TypeScript in production
- 5+ years designing backend platforms for production systems
- Strong systems design skills (boundaries, contracts, failure modes, scaling)
- Deep PostgreSQL knowledge (RLS, indexing tradeoffs, event sourcing, schema evolution)
- Strong application security architecture skills (encryption models, key management, auth patterns)
- Experience with policy-as-code authorization (Cerbos, OPA, Cedar, or similar)
- Strong code review skills for security-sensitive AI-generated code
- Experience with cryptographic primitives in production (digital signatures, hash chaining, app-layer encryption)
- Experience with event streaming platforms (Kurrent, Redpanda, Kafka)
- SOC 2 or GDPR compliance engineering experience
- Fintech or regulated-environment background
What We Do
Causa Prima is building the first agentic AI network for finance operations. Their autonomous agents move beyond simple task automation to negotiate across company boundaries, resolve disputes in real-time, and make financial decisions with real judgment. By leveraging a context graph of financial decisions, the company aims to create a scalable, acting system that optimizes cash and financial workflows for enterprises.
