Note: The job is a remote job and is open to candidates in USA. K3-Innovations, Inc. is seeking contractors for data intelligence engineering and computational drug discovery AI model development projects. The roles focus on designing scientific data pipelines and database structures, ingesting and integrating computational drug discovery results, and developing, fine-tuning, automating, and benchmarking machine learning models for protein-ligand structure and affinity prediction.
Responsibilities
- Design and create database tables to support a model inventory, including ontology, model metrics, and model results
- Enable ingestion and organization of results from a range of sources, including:
- Custom machine learning models
- Co-folding affinity prediction results
- Custom MPO calculations
- Other custom scientific calculations
- FEP+ results or related physics-based scoring
- Facilitate model containerization and deployment on the AID self-service platform to enable automatic API deployment
- Build data ingestion workflows for result integration, including use of staging tables to manage inserts and updates of new data
- If time permits: Develop scripts to enumerate virtual molecules using Free-Wilson or MMP transformation and compute associated predictions
- Evaluate open-source models and methods as needed to support project goals
- Start from a pre-trained foundation model such as Boltz-2 or AISB and fine-tune it on data
- Compile datasets from multiple advanced LO projects including both experimental protein-ligand complex structures and potency/affinity data
- Evaluate dataset variation across projects to challenge model applicability domains, ranging from small sets with a few structures and ~100 affinity data points to larger sets with 10+ structures and thousands of affinity data points
- Benchmark model performance against established methods and tools, including DeepAutoQSAR, GatorAffinity, MPNN, MM-GBSA, FEP, and related approaches
- For 3D workflows, compare pose generation using co-folder approaches against docking-based methods and assess performance
- Use project-specific thresholds to evaluate recall and precision as primary metrics
- Using confidence and domain of applicability, develop model ranking across affinity prediction model types
- Determine whether a single fine-tuned co-folder model trained across selected projects is sufficient, or whether project-specific models are required
- Design the workflow with automation in mind, ideally enabling an end-to-end process from protein and ligand sequence to predicted 3D structure and affinity
- Optimize workflow for efficiency that single predictions can be generated in less than one minute on GPU or CPU resources
Skills
- Python, SQL and scripting
- Database table creation and maintenance
- Containerization and AWS EC2 environment (or similar) familiarity
- Familiar with scientific data and machine learning
- Experience Level = 5-7 Years
- Strong experience in data engineering, scientific data management, or computational chemistry/cheminformatics environments
- Proficiency with relational database design and schema development
- Experience creating and maintaining staging, integration, and data pipelines
- Working knowledge of machine learning models, model metadata, and result tracking frameworks
- Familiarity with containerization and deployment workflows such as Docker and API-based model serving
- Experience with scripting and automation, preferably in Python
- Familiarity with cheminformatics concepts such as Free-Wilson analysis, matched molecular pairs (MMPs), and virtual molecule enumeration
- Ability to evaluate open-source tools and models for scientific use cases
- Strong collaboration and communication skills for working across scientific and technical teams
- Strong experience in machine learning, computational chemistry, cheminformatics, or related scientific data science domains
- Experience in 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
- Python and/or related scripting and automation experience
- Strong experience in machine learning, computational chemistry, cheminformatics, or related scientific data science domains
- Experience in training and fine-tuning models on domain-specific scientific datasets
- Work of benchmarking methodologies and model performance evaluation
- Experience with 3D modeling workflows, pose generation, or docking-based comparison studies
- Strong scripting and automation skills, preferably in Python
- Ability to design reproducible model workflows with attention to runtime and scalability
- Strong collaboration and communication skills for working across scientific and technical teams
- Experience supporting drug discovery or CDD-related data workflows
- Familiarity with FEP+ or related computational chemistry methods
- Exposure to ontology design and scientific data standardization
- Experience with cloud or platform-based self-service deployment environments
- Ability to work independently and deliver high-quality technical solutions in a contractor setting
- Experience supporting drug discovery or lead optimization projects
- Familiarity with co-folder models, Boltz-2, AISB, DeepAutoQSAR, GatorAffinity, MPNN, MM-GBSA, or FEP-related methods
Benefits
- Remote work arrangement
- 4-month contract (September–December)
Company Overview
Company H1B Sponsorship