Luana Martins
Pesquisa Em Aprendizagem Por Reforço A Partir Do Feedback Humano at Inteligência Artificial & Deep Learning Brasil
Goiás, Brazil
Summary
👤
Senior🎓
Top SchoolLuana Martins is a research scientist and MSc in Computer Science with 8 years of experience applying deep reinforcement learning and recommendation systems to real-world problems across agro, finance and energy sectors. Based in Goiás, Brazil, she leads research on human-in-the-loop RL for large language models, exploring the feasibility of using LLMs as evaluators to reduce experimental costs and scale feedback-driven training. Her portfolio spans production-oriented P&D: debt collection orchestration, personalized marketplace and tourism recommendations, intelligent vehicle inspection, and credit-recovery algorithms that leverage offline and multi-task RL for robust transfer. She has a track record of balancing rigorous academic inquiry with applied engineering—developing fair exposure recommendation algorithms and large-language solutions for the electric sector. Known for making RL generalize across tasks and detect out-of-distribution samples, she brings both classroom teaching experience and industry collaboration to bridge research and deployment.
9 years of coding experience
Mestrado em Ciências da Computação, Computer Science, Mestrado em Ciências da Computação, Computer Science at Universidade Federal de Goiás
English, Spanish, French, Portuguese
Github Skills (9)
meta-learning5
python4
ieee3
raycasting3
cpp2
simulation1
physics1
simulator1
quaternions1
Programming languages (5)
Github contributions (5)
AKCIT-RL/mujoco_playground
Mar 2025 - Jul 2026
An open-source library for GPU-accelerated robot learning and sim-to-real transfer.
Contributions:1 PR, 18 pushes, 6 branches in 1 year 3 months
AKCIT-RL/CORL
Feb 2025 - Jun 2026
High-quality single-file implementations of SOTA Offline and Offline-to-Online RL algorithms: AWAC, BC, CQL, DT, EDAC, IQL, SAC-N, TD3+BC, LB-SAC, SPOT, Cal-QL, ReBRAC
Contributions:10 PRs, 49 pushes, 5 branches in 1 year 4 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.