Open to opportunities

adriano bavieracv

@adrianobavieracv

Senior Full Stack Engineer specializing in AI/LLM system evaluation, RAG pipelines, and prompt-driven document workflows.

Brazil

Message

What I'm looking for

I’m looking to apply my evaluation-driven mindset to AI Trainer/RLHF and Model Evaluation—building reliable prompts, RAG/LLM integrations, and rubric-based quality systems that measure accuracy, relevance, and grounding at scale.

I’m a Senior Full Stack Engineer with 10+ years of experience building and evaluating AI/LLM-powered systems—especially RAG pipelines, LLM API integrations, and prompt-driven document workflows. I bring a hands-on evaluation mindset across accuracy, relevance, and quality, combining prompt engineering and model behavior with supervised machine learning approaches like XGBoost and Isolation Forest.

In recent work, I integrated LLM-based document summarization via the OpenAI API, cutting manual review time by 70%, and designed and tuned a RAG pipeline using LangChain and pgvector across 1M+ documents, improving retrieval accuracy by 34%. I’ve also led structured evaluation and quality review—training and validating ML classifiers with precision/recall-style analysis and implementing rubric-based code review and data-quality frameworks that reduced production defects by 45% and data-quality incidents by 70%—and I’m now applying this evaluation-driven mindset to AI Trainer, RLHF, and Model Evaluation work.

Experience

Work history, roles, and key accomplishments

Senior Full Stack Engineer

Kainos & Co.

May 2024 - Dec 2025 (1 year 7 months)

Integrated LLM-based document summarization via the OpenAI API, cutting manual review time by 70%. Developed a RAG pipeline with pgvector and LangChain across 1M+ documents (34% retrieval accuracy improvement) and built full-stack, API-driven data platform features using React/TypeScript and Node.js/Express.

OpenAI API Summarization Retrieval Augmented Generation (RAG)LLMs (LangChain Pgvector React TypeScript Node Express REST GraphQL Apollo JWT RBAC PostgreSQL Microservices Spring Boot fastAPI GitHub Actions CI CD

Senior Full Stack / Backend Engineer

Deloitte / Sow

Aug 2021 - Jul 2023 (1 year 11 months)

Developed an XGBoost predictive churn model (AUC 0.89) contributing to $1.2M in annual savings, and built a real-time fraud detection pipeline using Kafka and Spark Streaming (42% reduction in fraudulent transactions). Built an enterprise portal with Angular 14 and .NET Core 6, delivered 30+ REST APIs for SAP SD and Dynamics 365 integrations, and performed SQL performance tuning and zero-downtime

Python XGBoost Real Time Fraud Kafka Angular 14.0+.NET Core REST APIs SAP Integration Dynamics 365 Integration SQL Server Log Shipping

Data Engineer / Full Stack Developer

TaxPricing

Nov 2020 - Jun 2023 (2 years 7 months)

Created an Isolation Forest anomaly detection model in Python for tax fraud identification, mitigating audit risk by 35% and saving $500,000 in potential penalties. Built reconciliation and compliance solutions with Python, SQL, React, and FastAPI, and developed scalable ETL pipelines with SSIS and Azure Data Factory.

Isolation Forest Python SQL Anomaly Detection Data Validation React fastAPI ETL SSIS Azure Data Factory SQL Server Execution Plan Analysis

Full Stack Engineer

BASF / DecisionGroup

Jan 2020 - Jul 2020 (6 months)

Constructed a price elasticity model using Python and scikit-learn, improving profit margins by 12%. Launched a dynamic pricing portal with Angular 10 and Java Spring Boot, migrated 4 TB of data to AWS RDS using AWS DMS, and implemented REST APIs using OpenAPI 3.0 while optimizing performance with Redis caching.

Price Elasticity Python scikit learn Java Spring Boot AWS RDS Redis OpenAPI REST APIs

Software Engineer — Backend / Data

Google

May 2015 - Jun 2017 (2 years 1 month)

Founded and implemented a data quality framework with systematic validation and anomaly detection, reducing data-related incidents by 70%. Developed Java and Go backend services handling 50M+ daily API requests, built ETL pipelines using Apache Beam and Cloud Dataflow, improved BigQuery data models to reduce query costs, and built a real-time recommendation engine using Apache Beam and Bigtable.

Java Go Data Quality Anomaly Detection Apache Beam Dataflow BigQuery BigTable ETL Backend Services

Senior Backend / Data Engineer

Huawei Telecommunications

Jul 2010 - May 2015 (4 years 10 months)

Instituted coding standards and comprehensive code review processes, reducing production defects by 45% over two years. Deployed an Apache Flink-powered real-time anomaly detection system for billing transactions and architected high-throughput pipelines using Oracle GoldenGate and Kafka, supporting large-scale billing and analytics capabilities.