AI Lead

About the Role:You will serve as the technical backbone for AI innovation across molecular modeling, simulation automation, predictive ML, and generative design. This role blends the rigor of scientific computing with the architectural leadership required to scale enterprise-grade AI systems for global pharma customers.You'll mentor teams, architect ML infrastructure, and bring research-grade models into production with the reliability and elegance expected from a world-class deep-tech organization.Key Responsibilities:1. Technical Leadership & Architecture:- Architect end-to-end ML systems for molecular property prediction, protein-ligand scoring, ADMET forecasting, and generative molecular design.- Lead design decisions for GNN-based pipelines, transformer architectures, diffusion models, and hybrid physics-ML models.- Build scalable ML infrastructure that integrates computational chemistry engines, force-field computations, and cloud-based simulation workflows.- Define technical standards for data pipelines, model deployment, distributed training, and high-performance inference.2. Research Translation & Scientific Integration:- Collaborate closely with computational chemists, medicinal chemists, biophysicists, and scientific researchers to translate cutting-edge methods-- FEP+, MD, docking, QM/MM, ADMET models, free-energy calculations-into production-grade AI capabilities.- Validate ML architectures using scientific benchmarking datasets (FEP benchmarks, DUDE, ChEMBL, PDBBind, internal assay data).- Guide the integration of ML predictions into Schrdinger-style physics simulations, ensuring scientific robustness and reproducibility.3. ML Engineering & Deployment Excellence:- Lead the implementation of ML pipelines using Python, PyTorch/TensorFlow, CUDA, Ray, Kubernetes, MLFlow and cloud compute.- Oversee deployment of hundreds of ML models with strong monitoring, drift detection, explainability, and uncertainty quantification.- Build APIs, microservices, and data services enabling seamless integration with desktop scientific platforms, cloud applications, and partner ecosystems.4. Collaboration & Cross-Functional Enablement:- Mentor AI/ML engineers, data engineers, and applied scientists, guiding them on architecture, code quality, and scientific model design.- Partner with Product, Program, and Project Managers to define technical feasibility, plan execution, and break down complex scientific requirements into engineering roadmaps.- Work with enterprise teams to optimize GPU utilization, HPC clusters, cloud-native workloads, and multi-modal AI research workflows.5. Innovation & Strategic Impact:- Identify new methods and opportunities-GFlowNets, RL-based molecular optimization, protein generation models, multi-scale simulation AI accelerators.- Drive innovation initiatives to modernize scientific pipelines, enhance simulation accuracy, and cut compute cycles dramatically.- Participate in customer interactions, technical deep dives, and co-innovation discussions with large pharma partners.Required Qualifications:- 7-12 years of deep AI/ML engineering experience with at least 3+ years leading technical teams.- Strong expertise in computational chemistry, cheminformatics, structural biology, or scientific machine learning.Hands-on mastery of:PyTorch/TensorFlow, CUDAPython (scientific stack)GNNs, transformers, multimodal architecturesMLFlow, Airflow, Docker, Kubernetes, AWS/GCP/AzureDistributed training frameworksExperience deploying ML solutions in production at scale-preferably in scientific SaaS or enterprise platforms.Ability to dissect scientific problems and translate them into robust, elegant technical solutions.Preferred Qualifications:- Experience integrating ML models with simulation platforms (MD, docking, QM/MM, FEP).- Familiarity with molecular data structures (SMILES, SDF, MOL2, PDB) and chemistry toolkits (RDKit, OpenEye, DeepChem).- Background in physics-inspired ML, generative chemistry models, or protein structure modeling.- Strong publication or patent history in ML for drug discovery (bonus but not mandatory). (ref:iimjobs.com)