AI Engineer — Adaptive Learning Systems

Luca Ferretti

Full-time · Mid-level · Sao_Paulo

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We are an educational technology company based in São Paulo. We serve 3.4 million students across Brazil — from primary school to university preparation — through an adaptive learning platform that delivers personalised content, assessment, and tutoring. We are one of the largest edtech companies in Latin America by active users and we are growing. Our current AI stack handles content recommendation, learning gap detection, and performance prediction. It works. The generation of personalised explanations, hints, and learning pathways — the part that would make the most difference for students — is still largely rule-based and template-driven. We want to change that. We are hiring an AI Engineer to join the team building our next-generation adaptive learning engine, which will use large language models to generate personalised, pedagogically sound explanations and practice content calibrated to each student's current knowledge state. The work requires care: educational content for students aged 10–18 must be accurate, age-appropriate, and free from the hallucination and bias problems that make general-purpose LLMs unsuitable for direct classroom use without careful engineering.

Responsibilities

Build and evaluate LLM-powered explanation generation for the core subjects in our secondary school curriculum
Design and implement the RAG architecture that grounds model outputs in curriculum-aligned content and reduces hallucination
Develop evaluation pipelines for educational content quality — accuracy, age-appropriateness, pedagogical clarity — that run automatically before content reaches students
Collaborate with our team of curriculum specialists and learning scientists to ensure AI-generated content meets educational standards
Contribute to the safety and moderation infrastructure that prevents harmful or inappropriate content from appearing in student-facing applications

Requirements

4+ years of AI or ML engineering with hands-on experience building LLM-powered applications in production
Python for LLM integration, RAG pipeline development, and building the evaluation infrastructure that catches content quality issues before they reach students
Hugging Face for working with open-source language models, fine-tuning on educational content corpora, and running models on our own infrastructure where latency or cost makes API calls unsuitable
RAG pipeline design for grounding model outputs in our curriculum content and preventing factual errors in educational explanations
LangChain or LlamaIndex for orchestrating the multi-step reasoning pipelines that drive our adaptive tutoring flows
SQL for working with student learning state data and content performance databases
Portuguese language proficiency — the platform serves Brazilian students and the content is in Portuguese

Benefits

BRL 18,000 – BRL 24,000 per month (approximately USD $3,500 – $4,700)
Full remote within Brazil or compatible South American time zones
30 days annual leave — Brazilian labour standard
Health insurance (plano de saúde) and dental coverage
Platform access for your own learning or your children — not a joke, many of our team use it