Principal Software Architect

Abacus is scaling fast – more Data Solutions applications, more agent-assisted engineering, more complexity worth getting right. We're creating this role to put a single, senior technical voice at the center of that growth: someone who can take sharp architectural judgment and turn it into standards the whole organization can build on with confidence.

We're looking for an experienced Principal Software Architect to set the technical vision for how Abacus builds, secures, and scales its healthcare data platform. This is a senior individual-contributor role, not a management track – you'll be the architecture guru and evangelist for our platform, translating business and customer needs into target-state architectures across data, applications, cloud, and agent-assisted engineering, and mentoring engineers and Solution Architects as you go.

You'll build the reusable application foundation – design system, application blueprints, integration contracts, and production-readiness gates – that lets Data Solutions applications move fast without reinventing the wheel every time. And as agent-driven development becomes a bigger part of how we ship, you'll be the one turning architecture standards into skills, rules, templates, and evaluation rubrics that let AI-assisted output meet the same bar as our best human-written code – putting you at the center of how Abacus builds software for years to come.

You'll bring enough depth to guide decisions across data, application, integration, cloud, and agentic development, while staying pragmatic about fit-for-purpose architecture and delivery constraints. You won't do it alone – you'll partner closely with Product Engineering, our XPORT Principal Engineering team, AI Engineering, Security, and Client Solutions, bringing architecture to life from concept through production, not just on a whiteboard.

• Own and evolve the reference architecture for Abacus's core platform – data ingestion, transformation, distribution, and serving – partnering with teams where ingestion and distribution ownership is shared, and with Product Engineering on application and service layers.
• Set and enforce architecture standards for scaling, high availability/disaster recovery design, serverless patterns, and multi-tenant SaaS architecture across our AWS and Azure environments.
• Guide architecture for HITRUST and GovRamp compliance alignment, partnering with Security and Compliance so designs meet audit and control expectations from the outset rather than as an afterthought.

• Design scalable architectures for payer data solutions spanning claims, membership, provider, clinical, lab, and pharmacy/Rx domains, and their analytical and operational serving layers.
• Apply FHIR-based interoperability patterns and healthcare data standards where relevant to product or client implementations.
• Establish data contracts, lineage, quality controls, metadata, and integration patterns across ingestion, transformation, and serving layers.

• Own the design system and front-end/UX standards for Data Solutions applications, partnering with UX/Design on accessibility, branding, and consistent interaction patterns.
• Define standard user-management capabilities – tenant setup, roles, permissions, and admin experiences – as reusable patterns instead of something each team reinvents.

• Translate approved architecture standards into skills, rules, templates, and evaluation rubrics that our ADD Harness and agent-assisted development workflows can consume directly.
• Evaluate agent-generated code and designs against architectural gold standards, and build the scoring and review mechanisms that let us scale the quality of agent-produced merge requests across engineering teams.
• Partner with AI Engineering on semantic/context layers, agent interfaces (MCP and A2A), retrieval patterns, AI gateways, evaluation, traceability, and governed model access.

• Be an active senior member and contributor to Abacus's Architecture and Data Guilds – introducing ideas, taking them to realization, and helping drive a culture of innovation through hackathons and beyond.
• Mentor engineers and Solution Architects through pairing on designs, reviews, and pilots, growing architectural capability across the team rather than concentrating it in one person.
• Establish lightweight architecture review mechanisms that protect critical standards without slowing delivery, measuring adoption through signals like reuse, exception volume, and escaped design defects.

• Ensure every architecture package includes deployment topology, support model, rollback strategy, and operational telemetry, partnering with DevOps/Platform and Security on CI/CD, infrastructure as code, and observability. Help to establish quality metrics like DORA
• Hold designs to a resiliency bar appropriate for a platform of Abacus's scale, and help teams reason through trade-offs between delivery speed and operational risk.

In your first 6–12 months, you'll leave a clear mark on how Abacus builds software:

• Deliver a reference architecture for Data Solutions applications that teams adopt from day one – a shared blueprint, design system, API/data contracts, and starter implementations that let new work move faster instead of starting from scratch.
• Stand up the skills, rules, templates, and evaluation rubrics that let our ADD Harness and agent-assisted workflows produce work that meets our architectural bar – a foundation that scales with the team, not just with you.
• Build the architecture decision records, non-functional requirements, and operational-readiness checklists that give every release clear evidence it's production-ready – deployment, observability, support, and rollback accounted for before go-live.
• Grow architecture capability across the team through pairing, reviews, and ownership handoffs – leaving Abacus with more senior architectural judgment than it had when you started, not just your own.

• 15+ years of software engineering experience, including significant time in senior architecture roles designing and guiding complex cloud, data, integration, and application solutions from discovery through production operation.
• Demonstrated experience setting enterprise-level architecture standards and mentoring engineers or architects without formal management authority – Principal-level IC leadership, not a people-management track.
• Deep, hands-on experience with cloud platforms (AWS and/or Azure), including containers, Kubernetes, Git-based CI/CD, Infrastructure as Code (e.g., Terraform), observability, and secure production operations.
• Strong experience with data platforms and distributed data processing (e.g., Databricks, Snowflake, Spark) and with modern application and service architectures (Python, API-first services, React/TypeScript front-ends), identity and authorization, and multi-tenant SaaS patterns.
• Working knowledge of agentic and AI-enabled engineering – LLM-enabled applications, MCP/A2A , LangGraph or comparable orchestration, AI-assisted developer tools (e.g., Cursor), Agent Driven Development and Harnesses retrieval/context patterns, evaluation, and responsible AI controls.
• Understanding of healthcare payer, provider, clinical, and claims data ecosystems, or the ability to ramp quickly in a regulated data domain.
• Familiarity with HITRUST, GovRamp, or comparable compliance and control frameworks, and the ability to design for privacy, least privilege, auditability, and secure SDLC.
• Excellent communication skills – able to align senior business and technical stakeholders, explain trade-offs, and produce clear architecture artifacts for both technical and non-technical audiences.
• Bachelor's or Master's degree in Computer Science, Engineering, or a related field, or equivalent experience.

• Direct experience building or maintaining reference architectures, design systems, or developer platforms used by multiple teams.
• Experience defining architecture standards, rules, or evaluation rubrics for AI- or agent-assisted development workflows.
• Experience with FHIR-based interoperability or other healthcare data exchange standards.