Forward Deployed Engineer
From an unclear problem to a solution in production.

Co-founder · Solutions Architect & Backend Engineer

Custom software studio. Main client: AudenPass, a ticketing platform live in production with real payments.

1 · Scope and design

Gathered requirements with the client, wrote the spec with acceptance criteria and designed the backend architecture and data model (16 tables).

2 · Build it by directing AI agents

Built the backend (Node.js, TypeScript, Express, PostgreSQL, Redis) by directing AI coding agents (Claude Code with subagents) through TDD, criteria-to-test traceability in CI and review of every change: 50+ REST endpoints documented in OpenAPI and ~750 automated tests.

3 · Payments and stock

Integrated Mercado Pago payments (Orders API) with HMAC-verified webhooks, idempotency and periodic reconciliation from a background worker. Designed stock control with PostgreSQL transactions and locks so tickets can't be oversold across replicas; validated with a local k6 load test (400 concurrent users, zero oversold).

4 · Deploy and deliver

Deployed on AWS (Lightsail, SES, least-privilege IAM) with GitHub Actions: database snapshot before migrating, zero-downtime rollout and rollback. Defined an OpenAPI contract that let the frontend be built in parallel, and coordinated integration and acceptance testing with the client.

Functional Analyst / Solutions Engineer

Enterprise projects for clients in energy, government and public safety. I'm the first point of contact between the client and the development teams, and every project runs through the same cycle:

Clients under NDA; scope, volumes and names are not disclosed.

1 · Scope the problem

I sit with the client to understand the problem, then gather, analyze and document the functional requirements - for example, the requirements for an emergency-management solution for a national public-safety agency, organized by module with functional specs and process documentation.

2 · Design the solution

I turn that into a viable solution: architecture, relational data model, and integration of data from heterogeneous sources (KML, PDF, field forms). I present it to the client with demos and proofs of concept.

3 · Build it through to production

I share the approved design with developers and analysts and work with them until it ships - sometimes configuring the application end to end myself. Along the way I validate REST APIs and integrations with Postman, run functional tests, and automate repetitive work with Python.

Highlight · Oil & gas client

For an oil & gas client's field-operations platform, I designed a 3-level relational data model (report → event → follow-up) and a standalone inspection module, and configured the field application end to end myself, while the development team built the separate monitoring apps on that same model. One design held two independent implementations.