Skip to content
View pedrodicati's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report pedrodicati

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pedrodicati/README.md

Hey! I'm Pedro Dicati 👋

I'm a Staff Data Scientist at iFood, focused on building scalable AI systems that drive real business impact.

About Me

I'm a Lead Data Scientist and Applied AI Engineer with expertise designing and deploying machine learning and generative AI systems in production environments.

Currently at iFood, I work on ML systems that impact millions of users. Previously at Cognitivo.ai, I acted as a technical reference across multiple AI initiatives, supporting projects from architecture design to deployment and evaluation.

My work spans the full AI lifecycle:

  • Problem framing & solution architecture
  • Model development & experimentation
  • Production deployment & monitoring
  • Business impact analysis

Core Expertise

Generative AI & LLMs

  • Generative AI applications and AI copilots
  • Multi-agent systems and LLM orchestration
  • LangGraph, n8n, and complex workflow orchestration

Machine Learning

  • Recommendation systems and collaborative filtering
  • Pricing & elasticity-based ML models
  • NLP-based matching systems
  • ML deployment and APIs

Cloud & Infrastructure

  • AWS (certified) and Databricks
  • MLflow & model lifecycle management
  • Scalable ML systems

Tech Stack

Python | Scikit-learn | AWS | Bedrock | Databricks | MLflow | LangGraph | n8n | LLMs

Notable Achievements

🏆 Recommendation Systems - Built recommendation engines for fashion retail generating $300k–$400k in additional revenue with 3% margin improvement

🏆 AI Copilots - Designed multi-agent systems for customer service operations in education sector

🏆 Pricing Models - Developed ML pricing models for pharmaceutical distribution improving margins by 0.5pp

🏆 Generative AI - Built systems transforming unstructured marketplace data into structured product information using LLMs

Featured Projects

  • AI Copilot for Customer Service - Multi-agent architecture with specialized agents for call center support
  • Fashion Retail Recommendations - Collaborative filtering, product similarity, and basket recommendations
  • Pharmaceutical Matching & Pricing - NLP-based matching with demand elasticity pricing models
  • Generative Data Structuring - GPT-powered transformation of unstructured data into structured formats

Connect

LinkedIn · Email

Popular repositories Loading

  1. pedrodicati pedrodicati Public

    Config files for my GitHub profile.

  2. Solutions_Beecrowd_URI Solutions_Beecrowd_URI Public

    Troubleshooting codes in Beecrowd Judge

    C

  3. html-css-studies-curso-em-video html-css-studies-curso-em-video Public

    Estudos de HTML e CSS apoiados nas aulas da plataforma Curso em Video do Gustavo Guanabara

    HTML

  4. Projeto-Robo-POO Projeto-Robo-POO Public

    Projeto de um jogo de robôs desenvolvido para obtenção de nota no comporente de Programação Orientada a Objetos, implementado em python.

    Python

  5. jogo-adivinha jogo-adivinha Public

    Programa feito com intuito de simular um jogo de adivinhação, onde é sorteado um número e o usuário deve tentar acertar o número

    C

  6. teste-newm teste-newm Public

    Projeto desenvolvido para teste de estágio em desenvolvimento na empresa NewM

    PHP