João V.

Key Skills

Software

Other

AWS SageMaker

Top Subject Matter

Real estate content generation and conversational AI

Customer service conversational AI and dialogue modeling

Fintech customer support chatbots with RAG and monitoring

Top Data Types

Text

Document

Top Task Types

Fine-tuning

Entity (NER) Classification

Classification

Freelancer Overview

AI Developer at ProplexaAI (Intelligent content generation using OpenAI APIs; hybrid pipelines and evaluation-driven ite. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and AWS SageMaker. Education includes Master of Science, UNOPAR – Universidade Norte do Paraná (2025) and Bachelor of Science, UNOPAR – Universidade Norte do Paraná (2023). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning, Evaluation, and Rating.

Labeling Experience

AI Developer - ProplexaAI

AWS SageMakerText

Built production AI features using OpenAI GPT APIs to support intelligent content generation and user-assistive real estate workflows. Designed prompt engineering strategies to improve response accuracy, contextual relevance, and interaction quality. Created hybrid pipelines combining rule-based logic with LLM outputs to deliver reliable, scalable, domain-consistent experiences in production. • Integrated OpenAI GPT series APIs for RFP creation and proposal summarization • Implemented hybrid AI pipelines blending deterministic rules with LLM reasoning • Added AI-assisted search and filtering to improve discovery and engagement • Partnered with product stakeholders to define evaluation metrics and improve performance using real user feedback.

2025 - Present

AI Developer at ProplexaAI (Intelligent content generation using OpenAI APIs; hybrid pipelines and evaluation-driven iteration)

OtherTextFine-tuning

Built and optimized conversational and content-generation workflows using LLMs and prompt engineering to improve response quality and domain relevance. Designed hybrid pipelines that combine rule-based logic with LLM outputs to ensure reliability and consistency in real estate workflows. Collaborated on evaluation metrics and iterative improvements based on user interactions. • Automated RFP creation and proposal summarization workflows • Implemented prompt strategies for contextual relevance and interaction quality • Integrated AI-assisted search and filtering to improve discovery and engagement • Defined evaluation metrics and used feedback loops to refine model performance

2025 - Present

AI Developer - ItsyouAI

AWS SageMakerTextClassification

Architected and developed a conversational AI platform (ReVI VLA Bot) to deliver human-like natural language responses at scale. Integrated Mistral AI 7B with fine-tuning pipelines and built RASA dialogue management for intent classification, entity extraction, and multi-turn conversations. Leveraged OpenAI GPT APIs for fallback generation, summarization, and hybrid retrieval-augmented generation workflows with prompt and evaluation strategies. • Built RASA dialogue flows for robust customer query handling • Implemented hybrid RAG workflows combining retrieval and generation • Designed evaluation frameworks to benchmark model outputs against human-quality standards • Collaborated with product and QA teams to iterate model behavior based on feedback and metrics.

2025 - 2025

AI Developer at ItsyouAI (ReVI VLA Bot with fine-tuning, RAG, and RASA dialogue flows)

OtherTextFine-tuning

Architected and developed an end-to-end conversational AI platform with dialogue management, retrieval-augmented generation, and fallback generation. Integrated a Mistral open-source model with custom fine-tuning pipelines to improve latency and domain relevance. Built and maintained RASA dialogue flows using intent classification and entity extraction for multi-turn conversation handling. • Fine-tuned Mistral 7B using custom pipelines for response quality • Implemented intent classification, entity extraction, and multi-turn dialogue with RASA • Used OpenAI GPT series for fallback, summarization, and hybrid RAG workflows • Created evaluation frameworks to benchmark outputs against human-quality standards

2025 - 2025

Conversational AI Engineer at CloudWalk, Inc. (Production deployment and evaluation-driven retraining)

OtherText

Designed and deployed LLM-powered chatbots for WhatsApp and Slack for a fintech customer base. Refined RAG pipelines to improve retrieval accuracy and low-latency performance in high-volume support scenarios. Monitored production model performance with observability tooling to detect drift and trigger retraining workflows for SLA compliance. • Integrated OpenAI and Anthropic APIs with internal knowledge bases • Built context-aware multi-turn dialogue systems for customer support • Developed and operated FastAPI microservice integrations on AWS • Used drift monitoring and performance signals to drive retraining workflows

2024 - 2025

Show more

Education

Work History