Job Description

About
A well-funded, pre-Series A Platform Engineering startup building the
infrastructure layer that AI agents run on top of — a bi-temporal
organizational knowledge graph that structures and serves organizational
memory with provenance-attached facts, not guessed-at embeddings. SOC 2 Type
II and ISO 27001 already achieved; a hands-on CTO and a tight core team
setting foundational architectural precedent.

The Role
This is end-to-end ownership of the backend platform: the ingestion pipelines,
the graph write path, the REST and MCP API surface consumed by LLM agents, and
the multi-tenant AWS infrastructure underpinning air-gapped, isolated-VPC, and
cloud deployment modes. Decisions made here will define the platform's
architecture for years. You'll work directly alongside the CTO.

What You'll Do
• Design and own high-throughput ingestion pipelines that normalize
heterogeneous streams from Slack, GitHub, Google Calendar, Jira, HubSpot, and
HR systems — resolving entity semantics and ontology mapping at write-time
against a per-org ontology, not at query time.
• Build and maintain the bi-temporal graph layer: every fact carries valid-
time and transaction-time; old facts are invalidated with full provenance, not
deleted — a fundamentally different model from standard event-log or vector-
store approaches.
• Own continuous cross-system identity resolution at streaming cadence:
unifying the same person across HubSpot, Gmail, Slack, and a dozen other
integrations under a confidence-scored, graph-native entity model.
• Maintain low-latency MCP and REST APIs serving concurrent LLM agent
workloads with scoped access, rate limiting, and correctness guarantees — and
debug real concurrency problems (Redis race conditions under live ingestion
load) when they surface.

What We're Looking For
• 5+ years of Python backend engineering with production distributed systems
and event-driven architectures.
• Hands-on experience with graph databases or knowledge graph systems (Neo4j,
property graphs, or equivalent).
• AWS at scale (S3, Lambda, EC2, VPC, IAM) and comfort debugging concurrency
issues in Redis or similar under production pressure.
• Experience operating multi-tenant SaaS with compliance constraints (SOC 2,
ISO 27001, or equivalent).

Offer
$190K – $300K base + meaningful equity at pre-Series A stage. Remote-first
with New York as the team hub.