Product Engineer · AI-enabled products · Frontend architecture

Real products with AI.
Built to hold up.

Seventeen years shipping web and mobile software. I turn ambiguous workflows into working products, using AI where it creates leverage and engineering judgment everywhere else.

Experience
17 years across web, mobile, and product systems
Commercial proof
Parker onboarding measured in elapsed time
Current build mode
AI-enabled products with deterministic evidence
Demo access
3 product destinations linked

01 / Commercial proof

Product ownership measured in elapsed time.

Owned frontend delivery for onboarding and activation across enterprise and standard customer segments.

Parker

Growth Engineer, Product Frontend

Sep 2022–Dec 2024 via Lumenalta · May 2025–May 2026 direct

Enterprise onboarding

~30 days~1 week

Standard onboarding

~7 daysSame day
  • Activation experiments with Product, Design, and Data
  • Heap and LaunchDarkly instrumentation
  • Strategy and sequencing for a major Next.js upgrade
  • Business analytics for LTV, CAC, and time to value

02 / Selected work

Products that show the whole decision.

AI-enabled coaching, an operational B2B workflow, and an agent reliability lab—each with seeded scenarios, failure behavior, verification, and explicit production boundaries.

AI-enabled coaching product

LeagueLoop

Validated locally

Turn one finished match into one better decision next game.

A post-match coaching product where deterministic code extracts and validates the evidence around one bounded model-generated review.

  • Structured validation
  • Explicit paid-call control
  • Local retention and deletion

Operational B2B product

Extra Work Capture

Portfolio demo

Preserve extra-work evidence before the context disappears.

A phone-first workflow that carries one request from field capture through requester acknowledgment, actuals, pricing, outcome, and a printable evidence packet.

  • Idempotent mutations
  • Append-only history
  • Visible failure recovery

Agent reliability lab

BoundaryLab

Portfolio demo

Only one architecture stayed useful under attack.

BoundaryLab holds eight synthetic support cases constant across three agent architectures, then shows how context, memory, review, and capability boundaries change safety and completion.

  • Bounded profile: 8/8
  • 0 attack successes
  • 0 false refusals

03 / Working approach

AI delivery with a human acceptance point.

AI is part of both the product surface and the delivery workflow. Responsibility for architecture, evidence, product boundaries, cost, privacy, and acceptance remains human.

  1. 01

    Frame the product job

    Name the user, commercial stake, inputs, decision, useful output, and most costly failure before choosing machinery.

    Workflow brief
  2. 02

    Divide responsibility

    Keep facts, permissions, state transitions, and business rules deterministic. Give AI one bounded interpretive task.

    System boundary
  3. 03

    Test the uncertain parts

    Use evaluation fixtures, structured outputs, failure scenarios, telemetry, and cost limits to make behavior reviewable.

    Validation record
  4. 04

    Keep acceptance human

    A person owns architecture, privacy, product quality, and the decision to release, merge, publish, or spend.

    Review gate

04 / Contact

If the work is close to the business, I want to hear about it.

Senior individual-contributor and hands-on lead opportunities, remote or in Montevideo.