Data Engineer (AI Infrastructure)

Tax Relief Advocates•Irvine, CA

8h•$110,000 - $150,000•Onsite

About The Position

Build the data systems and services behind the AI platform, transforming complex financial and case documents into structured, trustworthy data for AI, analytics, and operational systems. Help build real-time services the organization depends on, contributing to an existing platform with design documents, tests, and CI. The work is hands-on, spanning design through deployment. The role covers batch and real-time data pipelines, document processing using OCR and AI models, data modeling across relational and document databases and data lakes, and creating clean datasets for AI/ML teams. It also includes building and operating APIs and background services, maintaining CRM and third-party integrations, supporting real-time sales tools and secure partner data exports, migrating older services to the primary cloud where appropriate, and supporting internal web tools. The engineer writes design documents, contributes to architecture decisions, works with infrastructure-as-code and CI/CD, protects sensitive personal and financial information, and mentors junior engineers while improving engineering standards. The position is full-time, salaried, exempt, and on-site in Irvine, California.

Requirements

2–3 years building production backend or data systems, or equivalent depth owning production systems end to end, making design decisions, and handling incidents.
A degree in Computer Science, Engineering, or a related field, or equivalent practical experience.
Deep experience in at least one backend language; the team mostly works in Python.
Strong SQL and data modeling skills across relational and document databases, including understanding indexes, query plans, transactions, isolation, safe schema changes, and consistency across stores.
Experience building safely rerunnable and backfillable pipelines over structured and unstructured data, including JSON and PDFs, and understanding how to model data for analytics.
Production experience running LLM or OCR workloads, including handling variable outputs, validating structured outputs, measuring quality, and managing latency and cost as volume grows.
Understanding of model deployment and serving, including capacity and quotas, version pinning and evaluation, rollbacks, and fallback when a model or region is unavailable.
Understanding of rate limiting, including client handling of HTTP 429 responses, service quotas and backpressure, and rate-limit errors for LLM deployments.
Understanding of distributed systems and service contracts, including versioned and backward-compatible APIs, events, and schemas; timeouts, idempotent retries, delivery guarantees, and handling work that cannot succeed.
Production experience on a major cloud platform (AWS, Azure, or GCP), running services on serverless and container platforms.
Understanding of networking, DNS, private connectivity, identity, and permissions, and ability to read, change, and deploy infrastructure as code through CI/CD.
Understanding of OAuth 2.0, OpenID Connect, SSO, MFA, token issuance, validation, and expiration; familiarity with current practices such as short-lived credentials and service-to-service authentication without shared secrets.
Ability to design role-based, least-privilege access, scope permissions to resources, and enforce authorization on the server rather than only in the UI.
Understanding of user and service account lifecycles, including provisioning, role changes, offboarding, and regular access reviews.
Ability to protect data by keeping secrets out of code and personal data out of logs, and to treat customer and company information as strictly confidential.
Clear communication through design documents, pull request descriptions, and incident write-ups.
A thoughtful engineering approach: framing problems, weighing options, documenting trade-offs, choosing simple workable designs, shipping tested and observable code, measuring before optimizing, and owning systems after release.

Nice To Haves

Experience with Python, Azure, React and TypeScript, or Google Cloud, where some older services still run.
Experience with orchestration and data platform tools such as Dagster, Airflow, or Databricks.
Experience with streaming systems such as Kafka.
Experience with retrieval and vector search.
Experience with distributed rate limiting or API gateway design.
Experience with CRM or third-party integrations with fragile authentication.
Experience in regulated or PII-heavy domains such as tax or finance.
Experience using AI coding agents in production work.

Responsibilities

Build batch and real-time pipelines that ingest data from documents, APIs, CRM systems, and internal services.
Build reliable, accurate, and cost-conscious document-processing systems using OCR and AI models.
Design and safely evolve data models across relational databases, document stores, and data lake storage.
Provide AI/ML teams with clean, structured, accessible datasets, including training data.
Build data-quality checks and monitoring to catch bad data before it is used.
Build and operate APIs and background services, handling slow, failing, or rate-limited dependencies.
Maintain CRM and other third-party integrations and repair them when they break.
Help build real-time sales tools, including lead routing and live caller lookup.
Contribute to secure, controlled data exports to external partners.
Maintain older services and migrate them to the primary cloud where appropriate.
Support internal web tools used by case and sales staff.
Write design documents and contribute to architecture decisions.
Work in and modify infrastructure-as-code and CI/CD systems as needed.
Protect sensitive personal and financial data in code, logs, configuration, and third-party integrations.
Mentor junior engineers and improve standards for code quality, CI, and releases, including guardrails for AI coding agents.