AI Platform Architect & Integration Lead

zerohash; City of Albany (NY); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Healthcare Insurance

Vision & Dental Insurance

Equity

Maternity & Paternity leave

WeWork All Access

WFH Yearly Stipend

L&D Yearly Stipend

Job summary

zerohash seeks a senior Python-first backend engineer to join the Applied AI team, building the access layer, audit logging, and system primitives for agentic commerce products. You will define schemas for credentials and actions, implement signed builds, and drive security-first architecture across the Core and per-department agents.

You will work on cross-institution interoperability, tamper-evident logging, and confidential computing while maintaining strong defense-in-depth controls and data

Qualifications

• Senior, Python-first backend engineer with production experience in event-driven systems.
• Experience composing LLM agents (Claude, GPT, or comparable) into standing systems with deterministic control planes.
• Security-first habits in regulated environments (MSB/VASP, bank, or comparable) with least privilege and read-only as defaults.
• Familiarity with capability-based authorization: attenuable tokens, workload identity, policy-as-code.

Responsibilities

• Own the access layer with native connectors and secured tokens while enforcing data residency.
• Define a versioned schema for signed credential and action records and ensure they can be cross-checked.
• Set up deployment pipelines, policy-as-code, and signed builds with SBOMs for auditability.
• Build the authority lease service with integrated kill switches and tamper-evident logging.
• Capture events into append-only, hash-chained logs for verifiable auditing.
• Migrate agents onto the Core and lead security reviews for read access across departments.
• Extend Core from read to bounded write with idempotency keys and time-locks for money moves.
• Design cross-institution primitives for interoperability with signed mandates and attestations.
• Develop self-hosted, zero-egress private stacks when needed by classification evidence.
• Evaluate state-of-the-art in agent frameworks, confidential computing, and zero-knowledge attestations.
• Maintain security-first posture by threat-modeling prompts, credential exfiltration, and deputy risks.

Skills

Python

Event-driven systems

API integration

CI/CD

LLM agents

Security-first

Capability-based authorization

Tools

Databricks

SBOM

Job description

zerohash seeks a senior Python-first backend engineer to join the Applied AI team, building the access layer, audit logging, and system primitives for agentic commerce products. You will define schemas for credentials and actions, implement signed builds, and drive security-first architecture across the Core and per-department agents.

You will work on cross-institution interoperability, tamper-evident logging, and confidential computing while maintaining strong defense-in-depth controls and data