Founding Lead Engineer / Principal Systems Architect

OpenTeams

1d•Hybrid

About The Position

We are building a confidential intelligent operations platform for evidence-governed analysis, operational reconstruction, model-assisted workflows, and high-integrity reporting in regulated domains. The first deployment focuses on healthcare integrity, provider-level identity mapping, licensing, ownership, source reconciliation, and defensible review workflows. We are seeking a hands-on Founding Lead Engineer / Principal Systems Architect to work directly with the concept architect and translate a large, complex system vision into production-grade software, data architecture, model integrations, validation harnesses, and secure Kubernetes-based deployment infrastructure. This is not a standard software engineering role. This is a founding technical role for building the core architecture of a serious AI/data platform from the ground up. The right candidate must be able to absorb abstract system concepts in real time and convert them into schemas, APIs, service boundaries, deployment artifacts, validation tests, and pragmatic engineering roadmaps.

Requirements

8+ years of professional software engineering experience, or equivalent exceptional experience.
Expert-level Python engineering.
Experience building production backend services, APIs, data pipelines, and distributed systems.
Strong SQL and relational database design experience, preferably PostgreSQL.
Experience with graph databases, knowledge graphs, or complex relationship modeling.
Experience with LLM integration, open-weight models, structured outputs, prompt/template management, or model-evaluation workflows.
Experience with Docker, Kubernetes, Helm, GitOps, CI/CD, and secure cloud or private infrastructure deployment.
Experience with data validation, audit logging, RBAC, secrets management, and secure software design.
Ability to design modular systems from ambiguous early-stage architecture.
Ability to translate non-engineering conceptual guidance into concrete software architecture and implementation plans.
Strong written documentation skills.
Comfort working directly with a non-engineer concept architect.
Comfortable translating analytical concepts into efficient production code, including: graph algorithms and centrality measures; entity-resolution and record-linkage logic; scoring systems and weighted evidence models; time-series and temporal-pattern analysis; recurrence or longitudinal-pattern analysis; statistical validation and benchmarking; performance optimization for large structured datasets.
Must be able to turn analytical concepts into practical, testable, and efficient software.

Nice To Haves

Experience with Nebari, Dask Gateway, Keycloak, or comparable data-platform infrastructure.
Experience with vLLM, KServe, Ray Serve, Ollama, llama.cpp, Hugging Face Transformers, or comparable model-serving infrastructure.
Experience with Neo4j, Cypher, graph analytics, graph ETL, or graph visualization.
Experience with OPA/Rego, policy-as-code, deterministic rule engines, symbolic validation, or explainable decision logic.
Experience with FastAPI, Pydantic, SQLAlchemy, Alembic, pytest, and modern Python service design.
Experience with Terraform, ArgoCD, Flux, Vault, Prometheus, Grafana, OpenTelemetry, or comparable DevSecOps tooling.
Experience with vector databases, hybrid retrieval, pgvector, OpenSearch, Elasticsearch, or comparable retrieval systems.
Experience with Dask, Spark, Kafka, Redpanda, RabbitMQ, or comparable distributed processing and event-streaming systems.
Experience in healthcare, government, legal, finance, cybersecurity, program integrity, or other regulated environments.
Familiarity with provider enrollment, NPI/NPPES, PECOS, LEIE/exclusion references, licensing records, corporate registries, or healthcare integrity workflows.
Familiarity with EDI healthcare transactions, eligibility files, managed-care encounters, FHIR, HL7, or EHR audit logs is helpful for later expansion phases.
Experience building AI systems with human review, auditability, evidence controls, and high-consequence output safeguards.
Experience with private-cloud, on-prem, hybrid, or air-gapped deployments.

Responsibilities

Work side-by-side with the concept architect to convert advanced system ideas into technical specifications, service maps, data models, APIs, schemas, tests, and deployment plans.
Translate verbal and written design guidance into architecture diagrams, implementation backlogs, acceptance criteria, and working prototypes.
Identify ambiguity, missing assumptions, engineering risks, security issues, and implementation conflicts.
Help turn an evolving concept architecture into reproducible, testable, maintainable software.
Build production-grade Python services, APIs, data pipelines, background workers, and orchestration logic.
Design clean service boundaries for ingestion, entity resolution, evidence management, review workflows, reporting, audit logging, and model integration.
Build deterministic, auditable workflows for high-consequence system operations.
Establish repository structure, coding standards, documentation practices, testing standards, and implementation discipline.
Design and implement relational schemas, graph models, object-storage structures, retrieval indexes, and audit records.
Build canonical identity and entity-linking systems that reconcile conflicting real-world records.
Support relationship topology, ownership mapping, provider-network analysis, and source-conflict preservation.
Implement data validation, source normalization, evidence linking, deduplication, and data-quality checks.
Build a model-agnostic adapter layer for open-weight and hosted models.
Implement multi-model routing for parsing, extraction, summarization, evidence explanation, report drafting, reviewer critique, and deterministic no-model workflows.
Integrate model-serving infrastructure such as vLLM, KServe, Ray Serve, Ollama, llama.cpp, Hugging Face, or equivalent tools where appropriate.
Implement structured outputs, prompt/template management, model-call audit, output validation, and model versioning.
Ensure model outputs remain constrained by evidence, rules, schemas, human review, and audit records.
Build rapid internal UI prototypes for evidence review, graph visualization, timeline inspection, review queues, report review, and audit inspection.
Design backend APIs and data contracts that allow a dedicated frontend or full-stack engineer to later build a production analyst/reviewer workspace.
Ensure human reviewers can inspect evidence, source conflicts, model outputs, rule triggers, and report language before high-consequence outputs are finalized.
Deploy services using Docker, Kubernetes, Helm, GitOps, CI/CD, RBAC, secrets management, observability, and secure environment practices.
Support cloud, private-cloud, hybrid, or OpenTeams/Nebari-aligned infrastructure where applicable.
Implement secure configuration, environment promotion, logging, backup/restore, and infrastructure-as-code practices.
Build deployment patterns that can support development, test, staging, and controlled pilot environments.
Build synthetic datasets, golden tests, regression tests, benchmark suites, schema tests, model-output checks, and security-boundary tests.
Validate ingestion throughput, entity-resolution accuracy, graph query performance, model latency, report generation, audit volume, and backup/restore behavior.
Ensure every major module has clear acceptance criteria and reproducible test evidence.