Formerly InfinomeBio, acquired by Inscripta in 2023

Develop machine learning models to address research/business problems. Work closely with collaborating biologists/managers to understand the relevant research questions. Also involved with mentoring/managing a small group of statisticians.Developed state-of-the-art causality based gene regulatory networks (GRNs) to understand how genes influence each other. Developed statistical & image-recognition based algorithms for automatic genotyping. Implementation of this algorithm will save $1 M annually.

Worked as part of a multi-disciplinary team to develop a novel catalytic technology to refine complex fluids such as bio-oil and FCC slurry oil. Developed theory-based machine learning models to estimate parameters for generalized Non-Random Two Liquid (NRTL) and Universal Quasi-chemical (UNIQUAC) models.Developed generalized machine learning models for predicting pure-fluid properties like surface tension, liquid viscosities, and vapor viscosities.Also involved in training/assisting team mates with modeling and general programming techniques.

Developed improved non-linear feature selection algorithms for semi-empirical QSPR models based on evolutionary algorithms and artificial neural networks.Implemented accurate machine learning models to predict octanol-water partition coefficient, melting points, boiling points and various other important physicochemical properties.Developed an improved framework for computer-aided molecular design (CAMD) and applied it to design new solvents for extractive distillation of cyclohexane and benzene, and to design new permeation enhancers for enhancing insulin permeation through the skin.Designed and performed in-vitro experiments to measure insulin permeability across porcine skin when exposed to different permeation enhancers.

Developed a novel method for synthesizing cost-optimal heat exchanger networks using evolutionary algorithms.