Core Architect & Integration Lead — Applied AI

zerohash•New York, NY

About The Position

zerohash’s Applied AI team works on an internal Intelligence Core—a system connectivity and control layer—and external agentic-commerce products. The Core Architect & Integration Lead is the engineering backbone for both, building the access layer, audit logging, skills and agents, and system architecture. The team builds the Intelligence Core itself; engineering maintains platform accounts, event hooks, and retrieval layers, while this role builds the systems above them. The role’s core objective is to connect company systems securely and create records of agent actions that third parties can verify. The internal build serves as a sandbox for agentic-commerce products, proving identity, authority, write permissions, and tamper-evident logging before they power multi-institutional agentic networks. This is a foundational role on a new team, with broad scope and a high degree of autonomy.

Requirements

Senior, Python-first backend engineering experience.
Production experience with event-driven systems, API integration, and CI/CD.
Experience composing LLM agents, such as Claude, GPT, or comparable models, and platforms such as Bedrock or comparable services, into standing systems.
Judgment to keep models at the edges and the control plane deterministic.
Security-first habits in a regulated environment, such as an MSB/VASP, bank, or comparable organization.
Treat least privilege and read-only access as default practices.
Working familiarity with capability-based authorization, including attenuable tokens, workload identity, and policy as code.
Ability to work autonomously in a foundational role with broad scope and limited hand-holding.

Nice To Haves

Experience with verifiable credentials, HTTP message signatures, or emerging agent-payment mandate schemes such as x402, AP2, MPP, Visa TAP, or Mastercard Agent Pay.
Experience with self-hosted LLM stacks.
Experience with confidential computing or trusted execution environments.
A research habit and autonomy.

Responsibilities

Own the access layer, including native connectors, MCP servers, and Databricks US/EU, using short-lived tokens, least-privilege read-only access, and no raw PII while enforcing data residency.
Define a versioned schema for signed credentials and action records in one namespace so each record can be checked against the other.
Establish build discipline, including deployment pipelines, authority policy as code, skill merge gates, and signed builds with SBOMs.
Build an authority lease service with integrated kill switches.
Capture events in append-only, hash-chained logs for tamper-evident, independently verifiable auditing.
Migrate existing team agents onto the Core, launch new per-department agents, and lead security reviews for continuous read access.
Extend the Core from read access to bounded write access, including logged actions, idempotency keys, compensating actions, and multi-approver time-locks for actions that move money.
Design core primitives for cross-institution interoperability, wrapping actions in signed mandates, scoped authority, and attestation receipts compatible with standards such as x402, AP2, and MPP.
Design self-hosted, zero-egress private stacks when required by classification evidence.
Evaluate and integrate developments in agent frameworks, confidential inference, and zero-knowledge attestations.
Threat-model prompt injection, credential exfiltration, and confused-deputy risks upfront, maintaining a security-first posture.

Benefits

Healthcare insurance: zerohash covers roughly 100% of employee premiums and a portion of spouse/children premiums (U.S. only).
Vision and dental insurance (U.S. only).
Chance to earn equity.
Maternity and paternity leave after six months.
WeWork All Access Membership.
Yearly work-from-home stipend.
Yearly L&D stipend after six months.

Stand Out From the Crowd

Upload your resume and get instant feedback on how well it matches this job.

Upload and Match Resume