Accelerating software engineering with quality, AI Engineering, and automation.
I'm Pedro Almeida, a Senior QA and Full Stack Engineer with 9 years of experience in system architecture, Web/Mobile test automation, AI Engineering, and Harness Engineering. Currently at Questrade Financial Group, focusing on test architecture for Design Systems and mission-critical mobile applications.
Pedro Almeida
Senior QA & Full Stack Engineer | AI & Harness Engineering
Salvador, Bahia, Brazil
Professional Summary
Integrated approach across full-stack development, distributed architecture, rigorous quality engineering, and AI-driven acceleration.
Full Stack Engineering & Software Architecture
Building end-to-end solutions in Go, TypeScript, React, Next.js, Node.js, and Flutter. Designing scalable microservices, concurrent telemetry and ingestion pipelines, relational persistence, and automated containerized pipelines with Docker.
AI Engineering, Harnesses & Continuous Quality
Architecting structured test harnesses, deterministic evaluation pipelines (evals), and agentic workflows for technical productivity. Implementing cross-platform automation frameworks (Detox, Playwright, Cypress, Jest) focused on system reliability and digital accessibility (a11y).
Professional Experience
Track record in major technology and financial services organizations.
Operating to drive documentation-driven QA practices across the Allspark Design System ecosystem (Web and Mobile) and the QuestMobile Explore module, aligning product, engineering, and support.
- Maintaining Playwright test suites and spearheading the migration to Cypress component testing with integrated digital accessibility (a11y) validation.
- Architecting end-to-end mobile automation with Detox for React Native applications in the QuestMobile Explore Module.
- Defining structured BDD test scenarios (@test-cases) and managing knowledge architecture in Confluence, ensuring strict alignment with Definition of Done (DoD).
- Delivering end-to-end traceability and test plan management integrating Zephyr, Jira, Confluence, and AI-assisted automation workflows (Atlassian MCP) for technical hand-offs.
- Enhancing Developer Experience (DevEx) via Dockerized container environments, GitLab CI pipeline orchestration, and automated Slack notifications.
- Providing Tier-1 operational support to Design System consumer teams (triage queues, distinguishing genuine defects from implementation issues) with published SLA and throughput dashboards.
- Applying AI Engineering and Harness Engineering (Cursor, Copilot, Claude, and context engineering) to accelerate technical documentation, test authoring, and organizational knowledge sharing.
Contributed to the shared data-flow standardization library bridging client applications and backend microservices across Pizza Hut/Yum! Brands.
- Developed automated test suites in Jest with state assertions on Redux store and React hooks.
- Engineered E2E test suites with Playwright and built backend services in Golang.
- Authored BDD test scenarios and orchestrated Docker test environments.
Engineered E2E and API test suites in Node.js environments integrated with Azure PaaS cloud infrastructure.
- Implemented Cypress and Puppeteer frameworks across the JavaScript ecosystem.
- Executed mutation testing, usability testing, integration testing, and BDD workflows.
Cross-functional role across Operational Support and Quality Assurance teams.
- Developed Python scripts for automated batch fixes in production environments.
- Automated test execution with Selenium and BDD (Cucumber and Ruby) with multithreading support.
Remote maintenance of telemetry and image capture devices, resolving data inconsistencies across SQL, Java, and Delphi systems.
Technology Matrix
Technologies, programming languages, and tools utilized daily.
Test Automation
Programming Languages
Frontend & Mobile
Backend & DevOps
AI Engineering, Harness & Quality Management
Projects & Applications in Production
Distributed systems, analytical platforms, and applications deployed and maintained in production environments.
Tô De Olho
Web · In Production · IFBA CapstoneWeb platform for legislative transparency and civic oversight dedicated to continuous monitoring of the Brazilian Federal Senate, developed as a Capstone Project at the Federal Institute of Bahia (IFBA). The system aggregates official open data on legislative bills, roll-call voting, committee engagement, parliamentary amendments, and quota expenses (CEAPS), computing a weighted parliamentary effectiveness and fiscal transparency index.
- Concurrent ingestion pipeline combining historical backfill routines with live synchronization against the Senate Open Data API and Transparency Portal.
- High-performance Go backend (Gin and GORM) with Redis caching and relational persistence in PostgreSQL 15.
- Decoupled Next.js 15 frontend (App Router), TypeScript, Tailwind CSS, Shadcn/UI, and analytical visualizations with Recharts.
- Docker Compose infrastructure with Cloudflare Zero Trust secure tunneling and continuous deployment.
Quem Votar
Web & Cross-Platform · In ProductionCross-platform civic application integrated into the Tô De Olho ecosystem, designed to provide public, transparent, and accessible insights into candidate profiles, asset declarations, and official electoral data from Brazil's Superior Electoral Court (TSE). The platform provides parameterized searches by office, state, and electoral cycle, advanced filters by registration status, party, and asset tiers, detailed breakdowns of declared assets and government plans, alongside a side-by-side comparison matrix of up to four candidates simultaneously without partisan bias.
- Reactive web interface compiled with Flutter Web utilizing BLoC state management and local persistence/caching via Drift (SQLite).
- Autonomous Node.js data worker with dedicated SQLite instance, executing scheduled 24-hour ingestion cycles across 109 electoral datasets from TSE DivulgaCandContas.
- Architectural decoupling ensuring third-party TSE infrastructure stays outside the critical path of user requests.
- Nginx web server on dedicated virtual machine (Contabo) with secure routing and Cloudflare edge caching.
NeuroAnálises
Web & Telemetry · In ProductionAutonomous platform for continuous telemetry, high-performance persistence, and time-series analysis of online gaming platforms (Double modality). Built for analytical auditing and probabilistic modeling, the application monitors real-time events, computing empirical frequency distributions, consecutive streaks, statistical skewness, and intervals between rare outcomes, enabling formal verification against theoretical multinomial expectations.
- High-cadence ingestion engine (7-second cycles) with exponential backoff, rate-limit resilience, and timestamp jitter compensation.
- Idempotent historical backfill with strict deduplication based on official round identifiers.
- TypeScript backend using Fastify and Better-SQLite3 in WAL mode, serving aggregated queries in under 50ms with less than 30 MB memory consumption in production.
- Real-time analytical dashboard with auto-synchronization (3s), dynamic time windows (1 hour to complete history), and batch data export in JSON and CSV formats.
MegaMoment
Android · Google PlayNative mobile application built with Flutter and published on the Google Play Store. Generates lottery numbers and draws based on the exact creation timestamp: lotteries, raffles, name pickers, and team balancers. Complete end-to-end execution: concept, development, cryptographic signing, release, and Play Console management.
- Native mobile development with Flutter framework and Dart language, following Material Design standards.
- Google AdMob monetization integration adhering to Google Play privacy policies and app-ads.txt compliance.
- Automated packaging, cryptographic APK/AAB signing, and full release management on Google Play Console.
Academic Background and Certifications
Rigorous academic foundation and continuous advancement in technology.
Academic Background
Key Certifications
Technical Publications
Academic contributions and published research papers.
Analysis of Wind Microgeneration Feasibility
Comparative feasibility study utilizing the City of Porto as an analytical benchmark for deployment in the City of Salvador.
Tô De Olho: Democratizing Senate Transparency through Open Government Data
Capstone Project approved at the Federal Institute of Bahia (IFBA), focusing on automated extraction, auditing, and public democratization of legislative open data for civic oversight.
View publication in IFBA repositoryGet in Touch
Available for technical discussions, mentoring, and professional opportunities.