Graph Neural Networks (Gnn) Specialist

Pubblicato il 17-11-2026 · Pharmanautix · Bardi

Pharmanautix Italia s.r.l., a company operating in the field of biotechnology research, is seeking a Graph Neural Networks (GNN) Specialist. The role involves periodic business travel within Italy and the possibility of occasional international business travel.
Contract Full-time, fixed-term contract
Classification C.C.N.L. Chimica e Affini – Industria, level D1
Compensation C.C.N.L. Chimica e Affini – Industria, level D1. Base monthly salary €2,****** over thirteen instalments — gross annual salary (R.A.L.) €26,******.
Location Milan, Italy — fully remote
Duration 12 months
What you will do
Expand our Graph Neural Network (GNN) engine.
Required
PyTorch Geometric.
Knowledge on training convolutional GNNs.
Hands-on experience in training GNNs for edge prediction tasks.
CUDA-accelerated ML.
Some understanding of biology or medicine.
Understanding of graph theory and representation learning.
PhD with focus on either computer science, bioinformatics, computational biology (or equivalent)



– OR master degree + 2 years working experience for a big tech company.
Some experience using cluster computing from remote access, e.g. SSH or similar.
Preferred
Knowledge on advanced GNNs, e.g. message-passing architectures.
Deep knowledge of biology.
Hands-on knowledge with transformers.
This job opportunity is open to candidates of either sex, without discrimination on grounds of gender, age, nationality or sexual orientation, pursuant to Italian Legislative Decrees No. *******, No. ********, No. ******** and No. ********, as amended. Personal data provided through CVs and the completion of forms will be processed in accordance with Regulation (EU) ******** (GDPR) and Italian Legislative Decree No. ********, as amended. The data controller is Pharmanautix Italia s.r.l. To exercise the rights afforded under Regulation (EU) ******** (GDPR), data subjects may write to ******
Full-time, fixed-term contract · Milan, Italy — fully remote
#J-*****-Ljbffr
Max. 8 MB. Formati Accettati: PDF, RTF, DOC, DOCX, ODT.