AI Engineer III – Computational Drug Discovery AI Model Development

Sam Quest Solutions Inc. is hiring an AI Engineer III to help build and refine AI/ML modeling capabilities for computational drug discovery. This role centers on developing, fine-tuning, and benchmarking models that move from protein/ligand sequence inputs to predictions of 3D structure and affinity, supporting multiple lead optimization programs.

In this remote position, you will design automated workflows and evaluate model performance across scientific datasets, with a focus on runtime efficiency and reproducible results. The work includes comparing alternative pose generation and docking approaches and determining how models should be structured for project-specific versus cross-project use.

What you’ll do

• Fine-tune pre-trained foundation models such as Boltz-2 and AISB using domain-specific project data
• Compile and curate datasets from multiple LO projects, including experimental protein-ligand complex structures and potency or affinity measurements
• Analyze dataset variation to stress-test model applicability domains
• Benchmark model performance against established methods including DeepAutoQSAR, GatorAffinity, MPNN, MM-GBSA, FEP, and related approaches
• Compare co-folder pose generation against docking-based methods
• Measure recall and precision using project-specific thresholds and build model ranking based on confidence and applicability domain
• Assess whether a single fine-tuned co-folder model can serve multiple projects or whether project-specific models are required
• Design an automated, end-to-end workflow optimized to produce single predictions in under a minute using GPU or CPU

What you bring

• Strong experience in machine learning, computational chemistry, cheminformatics, or related scientific data science domains
• Experience training and fine-tuning models on domain-specific scientific datasets
• Experience with 3D modeling workflows, pose generation, or docking-based comparison studies
• Working knowledge of benchmarking methodologies and model performance evaluation
• Strong Python skills, including scripting and automation
• Ability to build reproducible, scalable model workflows with runtime efficiency in mind
• Collaboration and communication skills across scientific and technical teams

Technologies

• Python
• GPU, CPU
• Boltz-2, AISB
• DeepAutoQSAR, GatorAffinity
• MPNN, MM-GBSA, FEP

Compensation and location

• Location: Remote
• Pay: USD 70 - 100 per hour