Advisor – Agent Research

Jobtailor; California (MO); On-site; USD 150,000 - 190,000; Full time

Job summary

Jobtailor is seeking a highly capable ML scientist to advance autonomous agents for molecule discovery. You will design reinforcement learning environments, engineer reward functions, and post-train models across chemistry and biology tasks.

You will integrate tools like RDKit, LIMS/ELN APIs, and instrument drivers while contributing to evaluation pipelines and MLflow tracking. The role involves collaborating with scientists, mentoring junior researchers, and presenting work to the AI community.

Qualifications

• PhD in ML, Bioinformatics, Cheminformatics, CS or related field or MS with 3+ years
• Hands-on collaboration with wet-lab or biology teams
• 1–2 years applying AI/ML to biology, chemistry, neuroscience, or related field
• Hands-on experience training or post-training AI models
• Proficiency in Python and ML frameworks
• Experience with RL and post-training methods (PPO, GRPO, DPO, reward modeling)
• Familiar with molecular representation learning or generative chemistry
• Experience building end-to-end cloud systems
• Cloud-native workflows (AWS/Azure, Nextflow, Argo on Kubernetes)
• Mentoring junior researchers or engineers

Responsibilities

• Partner with scientists to build autonomous agents for molecule discovery tasks
• Design and build reinforcement learning environments with proper state, action, and termination semantics
• Curate reward functions from noisy scientific signals
• Post-train domain models using SFT, DPO, GRPO, PPO, reward modeling, and distillation
• Integrate learned policies with domain tools (RDKit, ELN/LIMS APIs, instrument drivers)
• Build evaluation infrastructure, task suites, scoring harnesses, and experiment tracking (MLflow)
• Represent Frontier AI in the AI research community and publish work
• Evaluate external vendors, open-source projects, and academic collaborations
• Develop models and infrastructure to improve autonomous discovery and reduce DMTA turnaround
• Transition AI prototypes into production-deployed systems

Skills

Python

Reinforcement Learning

ML Frameworks

Post-Training Methods

Molecular Learning

Agentic AI Systems

Cloud Architectures

Mentoring

Publications

Education

PhD in Machine Learning

MS/BS with equivalent experience

Tools

PyTorch

TensorFlow

JAX

HuggingFace

MLflow

AWS

Azure

Nextflow

Argo

Kubernetes

Job description

• Partner with scientists to build autonomous agents for molecule discovery tasks
• Design and build reinforcement learning environments for real discovery tasks with appropriate state, action, and termination semantics
• Curate and engineer reward functions from noisy scientific signals
• Post-train domain models using SFT, DPO, GRPO, PPO, reward modeling, and distillation on chemistry and biology tasks
• Integrate learned policies with domain tools, including RDKit, molecular graph ML, ELN/LIMS APIs, and instrument drivers
• Build evaluation infrastructure, including task suites, scoring harnesses, regression tracking, and experiment tracking such as MLflow
• Represent Frontier AI in the broader AI@Lilly and external AI research community
• Publish research, give talks, review papers, and scout emerging trends
• Evaluate external vendors, open-source projects, and academic collaborations for strategic fit
• Develop models and infrastructure that improve autonomous scientific discovery and reduce DMTA turnaround
• Transition AI prototypes into production-deployed systems

Requirements

• PhD in Machine Learning, Bioinformatics, Cheminformatics, Computer Science, or related discipline, or MS plus 3 years / BS plus 5 years equivalent experience
• Demonstrated wet-lab collaboration or hands-on experience
• Approximately 1–2 years of experience applying AI/ML in biology, chemistry, neuroscience, or a related scientific field; industry postdoc counts
• Hands-on experience training or post-training AI models
• Proficiency in Python
• Deep experience with ML/deep learning frameworks such as PyTorch, TensorFlow, JAX, and HuggingFace
• Experience with reinforcement learning and post-training methods, including PPO, GRPO, DPO, reward modeling, RLHF/RLAIF, and libraries such as TRL or verl
• Familiarity with molecular representation learning, generative chemistry, or protein/nucleic acid models
• Hands-on experience building agentic AI systems
• Experience designing and shipping end-to-end systems in cloud environments
• Working knowledge of cloud-native AWS/Azure pipeline architectures, including Nextflow and Argo on Kubernetes
• Demonstrable research experience through project contributions and ideally publications in relevant ML/NLP venues
• Experience mentoring and guiding junior researchers or engineers

Core Competencies

Expertise in building and integrating autonomous agents for scientific discovery using reinforcement learning and machine learning frameworks. Proficient in Python and cloud-native architectures, with a strong background in collaborating with scientists and mentoring junior researchers.

Highest-signal resume keywords

• PhD In Machine Learning
• Proficiency In Python
• Experience With Reinforcement Learning
• Deep Experience With ML Frameworks
• Hands-On Experience Training AI Models

ATS Optimization Keywords

Hard Skills

• Reinforcement Learning
• Machine Learning
• Post-Training Methods
• Molecular Representation Learning
• Generative Chemistry
• Agentic AI Systems
• End-To-End System Design
• Cloud Environments
• AI Model Training
• Reward Modeling

Soft Skills

• Collaboration
• Mentoring
• Research Communication

Industry Keywords

• Bioinformatics
• Cheminformatics
• Molecule Discovery
• Scientific Discovery
• Neuroscience
• Chemistry
• AI Research
• ML/NLP Venues

Tools & Technologies

• PyTorch
• TensorFlow
• JAX
• HuggingFace
• MLflow
• AWS
• Azure
• Nextflow
• Argo
• Kubernetes