Gabriel Isaias Padua Carvalho

// skills

Languages · Cloud & Infrastructure

Languages: Python · PHP · TypeScript/JavaScript · SQL · Bash · C · C++

Cloud & Infrastructure: AWS (EC2, S3, Lambda, DynamoDB, API Gateway, CloudFront, WAF, ACM, Route 53, IAM, SES, Bedrock, EBS/DLM, GuardDuty) · Azure (AKS, Bicep) · Cloudflare (Tunnel, WAF, DNS) · Linux (Ubuntu, AlmaLinux, cPanel/WHM)

// skills

DevOps & Platform · AI & ML

DevOps & Platform: Docker · Kubernetes (k3s, AKS) · Terraform · Bicep · CI/CD & git-driven deploys · observability & monitoring (CloudWatch, UptimeRobot) · WireGuard · Tailscale · systemd

AI & ML: LLM production systems (Bedrock/Claude, GPT-4o-mini, Deepgram, Retell) · prompt & conversation-flow engineering · agentic & multi-agent systems · RAG · MLOps · model evaluation, guardrails & XAI · PyTorch · TensorFlow · Hugging Face · CNN/U-Net · Transformers

// skills

Data · Practices

Data: PostgreSQL/Supabase · MySQL · DynamoDB · REST APIs · ETL pipelines · pandas/NumPy

Practices: TDD (pytest, Vitest, Playwright, axe) · IaC · DevSecOps & security hardening · incident response & RCA · functional-core architecture · technical writing

// experience

Software & DevOps Engineer

WealthGoal Software (formerly part of JDL Strategies) · Gold Coast, QLD · 2024-12 – present

Companies split late 2025; Gabriel engineered the separation. Two-engineer team; contractor structure.

• Shipped a production voice-AI qualification agent — from a self-built AWS Bedrock (Nova Sonic) + Twilio prototype to a managed Retell platform — replacing the vendor's extraction with a PHP + GPT-4o-mini pipeline, live September 2025.
• Built and deployed a standalone multi-tenant Django support portal (Python 3.12, Supabase, AI-assisted ticket triage) as sole engineer — 604 automated tests and live client workspaces on a functional-core architecture.
• Delivered a production WhatsApp financial-qualification chatbot (FastAPI + GPT-4o-mini, 25-step conversation engine) in three days, with idempotent HMAC-validated webhooks, atomic opt-out compliance and a single-call prompt design that cut response latency ~50%.
• Migrated two production servers to new AWS accounts with row-parity-verified database cutovers, and executed the corporate infrastructure separation of the two companies — vhosts, three Cloudflare zones and registrar EPP domain transfers.
• Owned the AWS environment — WAF, three CloudFront distributions, ACM, GuardDuty and IAM — decommissioned a legacy stack saving ~US$18k/year, and rebuilt backups on EBS/DLM snapshots after finding the prior policy silently capturing none.
• Led estate-wide security hardening and incident response — a CSRF/SSRF/security-headers sprint across three servers (five adversarial QA rounds), a same-day SQL-injection scanner response, and an authorisation-gate content leak fixed across 163 files / 422 sites — alongside cPanel/WHM/Imunify360 and Exim/SES mail operations.

// experience

Casual Academic

Griffith University · Gold Coast, QLD · 2024 – present

• Teach across Object-Oriented Software Development, Mobile App Development and penetration-testing / ethical-hacking courses; lead tutorials and lab sessions.

// projects

gipc.dev — self-hosted operator platform

2026

Live-demo platform on bare-metal k3s behind a Cloudflare Tunnel (zero inbound ports), infrastructure-as-code in-repo; migrated the full domain + mail estate solo.

gipc.dev

// projects

Nina Nails — booking platform (client)

2026

Full-stack Next.js 15 / React 19 / Supabase product on Vercel (~15k LOC) with a three-tier test suite (Vitest, Playwright e2e, axe) and Google Calendar + email integration.

// projects

Seismic inversion (PyTorch U-Net)

2026

Reproducible U-Net (12.5M parameters) inverting 2,000 synthetic seismograms via a differentiable wave simulation; functional-core design, 27 deterministic tests.

// projects

Transformer financial-prediction platform

2023-2024

Node.js + Python orchestration to Azure Kubernetes; ETL over 3M+ records/day, sub-200 ms per query, ~12% MAE reduction, IaC (Bicep/Terraform), >90% coverage.

// projects

Real-time drowning detection (Griffith × Huizhou)

2024-2026

Edge computer-vision on Raspberry Pi/ARM at >=20 fps and 95%+ accuracy, with sub-100 ms multi-sensor fusion; published in IEEE Access.

// publications

Artificial Intelligence for Drowning Detection: Technical Feasibility Established, Operational Viability Unproven

G. I. P. Carvalho et al. · IEEE Access, vol. 14, pp. 97604-97628 · 2026-01

doi.org/10.1109/ACCESS.2026.3701060

// education

PhD, Computer Science

Griffith University · 2025 – 2028 (expected)

Adaptive, self-reconfiguring multi-agent AI for real-time decision-making; confirmation passed April 2026. Supervisors: Dr Larry Wen, Dr Qinyi Li.

// education

Master of Information Technology

Griffith University · 2024

High Distinction, GPA 6.63/7.0; Griffith Award for Academic Excellence (2023 & 2024).

// education

Bachelor of Laws (LLB)

Universidade Federal do Paraná, Brazil · 2018

High Distinction.

// certifications · awards · leadership

Certifications, awards & leadership

• ACS Skills Assessment — Software Engineer (261313), Developer Programmer (261312), DevOps Engineer (261316) (2026-04)
• IELTS Academic — Overall 8.0 (L 9.0), CEFR C1 (2025-12)
• ICPC — 3rd place, South Pacific regional, Level B division (2024)
• Research fellowship, Xidian University (Guangzhou) — AI, Big Data & Embedded Systems (6 weeks) (2025)
• Griffith Award for Academic Excellence (2023, 2024)
• Griffith Coding Club — Led weekly data-structures & algorithms workshops and team labs preparing peers for the ICPC.
• Griffith AI Research Colloquium (2026) — Presented early PhD findings on adaptive multi-agent systems.
• Research seed-funding — Contributed the partner-engagement workstream to a 2026 Griffith research funding application.