Sr. Software Engineer, Integrations

Technical Delivery: Own complex engineering tickets end-to-end, from design through implementation, testing, and production. Write clean, well-tested, maintainable code and hold that standard in review.

Technical Design & Architecture: Author and review Technical Design Documents (TDDs) and contribute to Architecture Decision Records (ADRs). Lead technical discussions within your domain and help the team make sound, well-documented trade-offs.

Integration Platform: Build and evolve our next-generation integration platform: SFTP, REST API, and S3 ingestion entry points; Step Functions preprocessing (file transfer, PGP decryption, validation); a compute-agnostic Transformation Engine that runs on both Lambda and Glue; and event publishing to downstream consumers and client webhooks. Drive the platform toward low-code, config-driven client onboarding.

Sensitive Data & Tokenization: Own and harden the services that keep PANs and PII protected: format-preserving tokenization, PGP key management and rotation, SFTP relays, and redaction, all running in an isolated PCI environment. Our hardest success metric is zero unencrypted cardholder data entering the platform.

Client Data Contracts & Infrastructure as Code: Map client file and API formats into our canonical internal data schema (OTDS), and manage schema versioning and compatibility. Treat infrastructure as a first-class deliverable: Terraform-based client provisioning and environments, alongside CDK for the services you inherit.

Legacy Systems & Migration: Operate and improve the established integration services that serve our largest clients, including a Kubernetes-based file-processing platform and dedicated client APIs. Help migrate clients off them safely. Reduce the long tail of under-documented inherited systems by turning tribal knowledge into runbooks and alerts with clear ownership.

Client Go-Lives: Partner with Implementations, Solutions Engineering, and operations teams to bring new clients live: integration setup, dry runs, cutover, and hypercare.

AI-Assisted Engineering: Use AI coding agents as a primary part of your workflow and improve how the team uses them. That means maintaining repo context and skill files, building and consuming MCP servers that expose our tools safely, and codifying investigation-heavy work (triage, file reprocessing, client data questions) into reusable agent workflows. Help design and build the LLM-backed services we are adding to our pipelines, including prompt and context design, structured output and tool calling, validation loops, and model selection. Manage token budgets and cost with the same rigor as any other infrastructure spend.

Reliability & Observability: Own the operational health of your systems and take part in on-call. Instrument pipelines with meaningful signals (file- and record-level failures, SLA breaches, DLQ depth, latency), and replace manual failure reviews with automated monitoring.

Mentorship & Technical Leadership: Support the growth of more junior engineers through code review, pairing, and direct feedback, including raising the team's effectiveness with AI tooling. Lead spikes and timeboxed discovery work within your domain.

Cross-Functional Collaboration: Partner with Data Services, Data Engineering, Data Science, and Product to define clear data contracts and coordinate schema changes, so that the client data we ingest is fit for downstream consumers.

Education & Experience: Bachelor's degree in a related technical field, or equivalent practical experience.

Track Record: 5+ years of software or data engineering experience, with demonstrated ownership of production systems. No people management requirement.

Technical Mastery: Deep experience building data ingestion or integration systems: file- and API-based pipelines, idempotency, retries, dead-letter handling, and reprocessing. Strong AWS fundamentals, especially serverless and data services (Lambda, Step Functions, Glue, S3, EventBridge, Transfer Family) plus IAM in a multi-account setup. Production-grade Go, and comfortable picking up other languages when inherited systems require it. Comfortable with Kubernetes and with Terraform at production scale.

Security & Data Handling: Experience handling PII, and ideally cardholder data, in a regulated environment: encryption, tokenization, key management, and least-privilege access. PCI DSS exposure is a strong plus. That judgment extends to what data reaches an AI model, and to treating prompt injection and over-broad agent permissions as real risks.

AI & Modern Tooling: Most of your daily work will be done with LLM coding agents (Claude Code), so you already use agentic tools as your main way of working and have grounded opinions on their limits. You understand MCP in practice: you have consumed MCP servers and ideally built one, and you know how to package recurring work into reusable skills and workflows. You have hands-on experience building something real on an LLM API (Bedrock, Anthropic, or similar): prompt and context engineering, structured output, tool calling, evaluation and guardrails. You reason about model tiers, token budgets, and dollars, and can defend the cost profile of what you build.

Ownership Mindset: You take initiative, follow through without close direction, and treat production reliability as a personal standard, not a checklist. You prioritize what the team and its clients need, and you're comfortable inheriting under-documented systems and making them better.

Communication: You write clear TDDs and design proposals. You can explain technical trade-offs to non-engineers, including client-facing partners, and you flag cross-team dependencies early.