Forward Deployed Solutions Architect

Recurring Decimal Inc.; Phoenix (AZ); On-site; USD 140,000 - 190,000; Full time

Job summary

Recurring Decimal Inc. is seeking a Forward Deployed Solutions Architect to lead end-to-end technical delivery for agent-based commerce trust systems.

You will define architecture, oversee integration with Anthropic, UCP, OpenAI ACP, and other partners, and drive ADRs, threat models, and runbooks for live demos. You will own the design of identity binding, gateway verification, loyalty signaling, and verifiable credential flows, while communicating complex concepts to merchants and external

Qualifications

• 8+ years in solutions/software architecture with hands-on code and recent production work on LLM agents.
• Working knowledge of Claude (Messages API tool use, prompt caching, Agent SDK, Managed Agents) and MCP, including authenticated MCP servers.
• Identity and payments experience: OAuth 2.0/OIDC, JWT/JWKS, HTTP message signatures, and PCI-adjacent checkout flows.
• Threat-modeling and privacy-by-design skills. You write ADRs engineers follow and sequence diagrams partners can read.
• Consulting-grade communication. You've led working sessions with external engineering teams and bridged executives and builders.

Skills

Solutions architecture

Software architecture

Claude API knowledge

MCP & authenticated MCP servers

OAuth 2.0/OIDC

JWT/JWKS

PCI checkout familiarity

Threat modeling

Privacy-by-design

ADR writing & diagrams

Consulting communication

Tools

Cloudflare Workers

Job description

Forward Deployed Solutions Architect - Agentic Commerce Trust

AI agents are starting to shop and pay on behalf of consumers. There is a need to build the trust layer that tells merchants who (and what) they're transacting with. You are expected to own the technical solution end-to-end: deciding where trust is enforced in each ecosystem, writing the ADRs, running the demos, and serving as the technical face to merchants and partners (Anthropic, Google/UCP, Cloudflare, Visa, wallet vendors).

First 90 days

• Build the hook-point inventory and runtime comparison for Anthropic's blueprint (Messages API vs. Agent SDK vs. Managed Agents). Author ADRs on primary runtime target and integration architecture.
• Design the reference integration: session-start identity binding, backend decisioning (approve / step-up / deny) with per-hook fail-open/closed policy, and a gateway verifier for hosted agents. Merchants change zero prompt bytes and zero gate code.
• Design loyalty-without-login on UCP (namespaced loyalty signal, member pricing, reserved inventory, step-up via identity linking) and on a protocol-agnostic header/middleware rail.
• Shape the Verifiable Credential POC: issuer, holder wallet, presentation flow, merchant verifier, and required privacy properties.
• Write threat models, merchant integration guides, and demo runbooks with fallbacks. Then run the demos.

Must have

• 8+ years in solutions/software architecture with hands-on code, including 2025-26 production work on LLM agents with tool use.
• Working knowledge of Claude (Messages API tool use, prompt caching, Agent SDK, Managed Agents) and MCP, including authenticated MCP servers.
• Identity and payments experience: OAuth 2.0/OIDC, JWT/JWKS, HTTP message signatures, and PCI-adjacent checkout flows. Familiarity with UCP, OpenAI ACP, AP2, and KYA.
• Threat-modeling and privacy-by-design skills. You write ADRs engineers follow and sequence diagrams partners can read.
• Consulting-grade communication. You've led working sessions with external engineering teams and bridged executives and builders.

Nice to have

• W3C VC 2.0, OID4VCI/OID4VP, SD-JWT or BBS+
• Fraud/risk decisioning platforms
• Cloudflare Workers / edge architectures
• Prior forward-deployed or FDE role

You’ll own