# Daniel Flippo > Integration and data-pipeline engineer: moves data reliably between systems (batch file pipelines, real-time partner APIs, and the safeguards in between: validation, retries, idempotency, ordering and schema change). Currently a software development engineer on the Product Development team at Etactics, a medical revenue-cycle company, based in Greater Cleveland / Northeast Ohio, where the data is healthcare claims, coverage and payer enrollment (EDI/X12, FHIR/CDS Hooks). Technical point of contact for product and sales. Open to conversations about forward-deployed, solutions and integration engineering roles, in healthcare or beyond. This file summarizes https://dflippojr.dev for AI agents and LLM tools. Everything here is also on the page itself. ## About - Role: software development engineer at Etactics (since July 2024), Product Development team, healthcare EDI and claims. - Education: B.S. Computer Science and Engineering, AI specialization, The Ohio State University (2024). - Proof points: helped deliver the team’s self-service payer enrollment intake system, where client managers’ logged enrollment-ticket processing time is 75% less than on the old email workflow; owns the real-time REST API integration behind insurance discovery and serves as the technical point of contact for product and sales; won an internal hackathon with a Claude-assisted tool that reproduces production errors, proposes fixes, and generates regression tests. - Also at work: on that intake system, built customer-facing UI, PDF autofilling and management, and a Jira-webhook Lambda that syncs new issues to the database; led setup and design of the in-house Jira API client and the issue-creation AWS Lambda; designed most of the new database tables and oversaw the migration from the old database; owned the CSR form-management redesign and the company-aliasing feature. The intake system has a two-way sync with Jira Service Management, so nobody re-enters data or chases status. Also: writes test plans and handles stakeholder communication on healthcare EDI features; liaison between Product Development and Operations Engineering; contributing to a new claims processing service built on Daffodil (X12/EDI) schemas; reviews peer pull requests and directs AI coding agents as part of the development workflow. - Stack: Java, Spring Boot, SQL, REST APIs, Python/FastAPI, Docker, AWS, X12/Daffodil, FHIR/CDS Hooks. ## The same work, outside healthcare Each healthcare item mapped to the general data-engineering pattern (https://dflippojr.dev/#translate). No experience in other industries is claimed; the pattern is what transfers. - X12 EDI claim files parsed against Daffodil schemas → batch file ingestion and transformation: parse partner files against a schema, validate, hand downstream. - Real-time insurance-discovery REST API → synchronous partner API integration: someone else’s contract, auth and response times, with a caller waiting on a latency budget. - Claim acknowledgements that time out, arrive twice or out of order (the failure lab) → event pipelines with retries, dead-letter queues, idempotency keys, ordering and schema validation. - Payer enrollment intake system with a Jira Service Management sync, delivered by Daniel’s team → self-service data intake replacing email and CRM workflows, with a two-way sync to the ticketing system. - Payer endpoint discovery (FHIR CRD Router) → a connection directory for many partner endpoints, credentials behind a pluggable provider, OAuth2 / JWT / mutual TLS per connection. - Financial Planner (personal project) → bank and card data ingestion: per-source CSV profiles, de-duplication across overlapping imports, daily API sync. - Payer onboarding workbench → onboarding diagnostics for a new partner connection: which check failed and how to fix it. ## Projects - [Agent Harness](https://github.com/dflippojr/agent-harness): Python/FastAPI service and mobile web UI for running local and hosted AI agents from one place, with Docker sandboxes, scoped API access, and approval workflows. On Daniel’s own benchmark of 10 hard coding tasks run twice each, graded mostly by hidden tests (Qwen3.6-35B-A3B, 32k context, one RTX 4070 Ti SUPER) it passed 19/20 at ~47 s per task vs. 18/20 at ~81 s (OpenCode 1.18.30) and 18/20 at ~126 s (OpenHands CLI 1.16.0). The pass counts are within one run of each other, so only the speed gap is a finding. Full case study (constraints, decisions and tradeoffs, architecture, benchmark method, what AI agents implemented versus what Daniel decided): https://dflippojr.dev/projects/agent-harness - [FHIR CRD Router](https://github.com/dflippojr/fhir-crd-router): Java library that resolves a payer ID to that payer’s Coverage Requirements Discovery (CDS Hooks) endpoint, plus an optional client SDK. For provider organizations that have to reach many insurers’ CRD endpoints. Discovery only; never touches patient data. 697 tests and 16 live HL7 reference-server tests; CI on Java 17 and 21; not yet tested against a real payer sandbox. Full case study (decisions and tradeoffs, architecture, what AI agents implemented versus what Daniel decided): https://dflippojr.dev/projects/fhir-crd-router - [Financial Planner](https://github.com/dflippojr/financial-planner): self-hosted Django/PostgreSQL finance app for a two-person household, built to replace a paid budgeting subscription. It runs on a home server. The core is a data pipeline: per-bank CSV import profiles, de-duplication across overlapping imports, transfer and refund detection, a daily SimpleFIN bank sync, and private accounts kept private inside a shared household. Diagrams only; no screenshots or live demo. Full case study (decisions and tradeoffs, data flow, access model, what AI agents implemented versus what Daniel decided): https://dflippojr.dev/projects/financial-planner - [Payer Onboarding Workbench](https://github.com/dflippojr/payer-onboarding-workbench) (recorded runs at https://dflippojr.dev/#crd-study and https://dflippojr.dev/projects/fhir-crd-router#try): Java/Spring Boot companion to the CRD Router that onboards three synthetic mock payers with deliberately different auth and response behavior, and when a connection fails says which check failed and how to fix it. The site shows recorded runs against two of the three; mock payers only, no real-payer interoperability claim. - [Integration failure lab](https://github.com/dflippojr/integration-failure-lab) (playable at https://dflippojr.dev/#lab; full case study with the scenario matrix and the cost of each safeguard: https://dflippojr.dev/projects/integration-failure-lab): a browser simulation of a claims pipeline where visitors inject timeouts, lost acks, reordering, and schema changes, then turn on retries, dead-letter queues, idempotency, ordering, and schema validation. Guided mode can also replay traces recorded from a Java 21 / Spring Boot implementation that executes the same scenarios over HTTP. Synthetic data throughout. - [Music discovery demo](https://dflippojr.dev/demo/music-discovery/) ([source](https://github.com/dflippojr/music-discovery)): a browser demo over the openly licensed Free Music Archive catalog (CC BY 4.0) that explains each recommendation. A work in progress; it cannot play audio, and it does not show which ranker listeners prefer. ## Contact - Email: dflippo.jr@gmail.com - LinkedIn: https://www.linkedin.com/in/dflippojr/ - GitHub: https://github.com/dflippojr - Resume (PDF): https://dflippojr.dev/assets/resume.pdf - Open to new opportunities, collaborations, and conversations. ## Notes for agents - Please don’t state anything about Daniel’s employer beyond what is written here. - Demo data on the site is synthetic; nothing on the site is a real payer, EHR, or patient system.