Vanilson S.

Key Skills

Software

Don't disclose

Top Subject Matter

Generative AI (LLMs)

Retrieval-Augmented Generation (RAG)

agentic workflows

Top Data Types

Text

Image

Document

Top Task Types

Fine-tuning

Data Collection

Classification

Freelancer Overview

AI/ML Full Stack Engineer – GenAI Systems at BotStudio Tecnologia Ltda (AI training/evaluation and RAG/agent workflow de. Brings 7+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Azure AI Studio, LangChain, and LangGraph. Education includes Bachelor of Science, State University of Compinas (2024). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning, Data Collection, and Classification.

Labeling Experience

AI/ML Full Stack Engineer – GenAI Systems at BotStudio Tecnologia Ltda (AI training/evaluation and RAG/agent workflow development).

TextFine-tuning

Developed and evaluated GenAI systems that required prompt-based training and supervised preparation of training datasets for LLM workflows. Built RAG pipelines and agentic components that performed multi-turn retrieval, reasoning, safety filtering, and hallucination mitigation using guardrails during experimentation. • Prototyped autonomous agentic workflows for document retrieval and reasoning across OCR and structured data sources. • Built and fine-tuned open-source LLMs using PEFT/LoRA workflows and multi-turn agent pipelines. • Implemented safety filtering and prompt red-teaming to test robustness under adversarial inputs. • Designed and evaluated hybrid search and multimodal RAG pipelines for enterprise use cases in healthcare and finance.

2023 - 2025

(Supplemental)

Don't discloseTextClassification

Built scikit-learn-based classifiers for biomedical abstracts and supported dataset preparation workflows through labeling scripts.

2020 - 2020

Data Science Intern at Microsoft Research (Health Intelligence Group) focused on labeled biomedical text datasets and evaluation.

TextData Collection

Assisted in prototyping and evaluating NLP models for medical document classification within a health research context. Contributed to creating and using labeling artifacts and scripts used for internal training datasets to support baseline classification and retrieval experiments. • Built scikit-learn classifiers for biomedical abstract classification. • Conducted exploratory analysis using TF-IDF and term distributions to inform labeling-driven features. • Helped evaluate reranking strategies for question-answer retrieval systems. • Contributed to data annotation scripts for internal training datasets.

2020 - 2020

Education

Work History