Hi! I'm Gabrielly, but you can call me Kimi 🧩
I'm a Machine Learning & MLOps Engineer with 2 years of experience. I specialize in high-scale data systems and human-centered AI that drives meaningful ROI.

I'm a penultimate year undergraduate Data Management student at the Federal University of Piauí (UFPI), Brazil (Current GPA: 9.47/10).

👩‍💻 Currently, I am...

• 💼 ML Engineer | Data Engineer at Cortex: Engineered an autonomous Agentic AI system for 80K+ stores, achieving 80-125% data quality and enabling $2M in revenue.
• 📊 Analytics Engineering Consultant (Freelance): Managed telemetry for 3.26 Billion requests and 67.67 TB of traffic during Black Friday 2025, maintaining 99.92% uptime.
• 🤖 AI Researcher at UFPI: Building hybrid BiLSTM-Transformer models for hate speech detection. Test the baseline
• 🕵🏼 On going undergraduate thesis: Researching Quantum Machine Learning vs. Classical architectures for NLP in Brazilian Portuguese.

🛠️ Tech Stack & Expertise

• Infrastructure & MLOps: Kubernetes (EKS), Karpenter (Autoscaling), Spot Instances, Docker, Terraform, MLflow, AWS (S3, Lambda, EC2, EKS, DynamoDB).
• Data Engineering: SQL (PostgreSQL, Supabase), Clickhouse, ETL Pipelines, Data Pipelines, Analytics Engineering.
• Machine Learning: LLMs, Agentic AI, Orchestration (Custom-built in Python and LangChain) NLP, Physics-Informed ML, TensorFlow, PyTorch, Vector Databases (RAG).
• Observability: HyperDX, Grafana, CloudWatch (Latência P95 & Error Budgeting).
• Programming: Python, C++, PySpark, R
• Core Math: Linear Algebra, Statistics, Probability, Calculus
• Quantum Computing: Qiskit, Pennylane (Research focused on QML for NLP)

✨ Achievements

• 🏆 Transpetro/Petrobras Hackathon: 7th place developing a Physics-Informed ML architecture for industrial prediction.Test the model
• 🇧🇷 I Speak Kanoê: Developed the 1st and only dataset for the Kanoê indigenous language preservation. Documentation | Download the Dataset
• 🏆 3rd Place at aTip Hackathon: Accessibility-focused AI solution. Project
• 📊 Social Impact: Created a unique Water Inequality Dataset of NE Brazil and published peer-reviewed research. Read the paper

🎖️ Certifications

• 🎖️ Stanford University and DeepLearning.ai: Machine Learning Specialization. ML Project - Credit Risk Prediction
• 🎖️ Harvard’s Aspire Institute: AI-Integrated Leadership Program (AILP) Alumna.
• 🎖️ DataCamp: AI Engineer Associate & Data Engineer Associate.

📝 Publications * 📄 ENUCOMPI (Peer Reviewed): Exposing water gaps and health risks in northeastern Brazil through data.

📚 My "Stanford Roadmap"

• As a neurodivergent learner, I thrive on structure and depth. I am executing a rigorous, self-directed curriculum based on Stanford University's Computer Science BS parallel to my formal degree.

🌍 Leadership & Volunteering

• 🧑‍🏫 Teaching Fellow: Data and AI for Social Analysis course at UFPI.
• 🧑‍🏫 AI Instructor: Volunteering at an NGO teaching AI for women who wanted to join the tech industry.
• 🇺🇳 United Nations MGCY: Volunteer at the Major Group for Children and Youth.

💬 Let's connect

• 💼 LinkedIn: Gabrielly Gomes
• 📧 Email: gabrielly.gomes@ufpi.edu.br