Rohola Zandie
CTO and Co-founder at Absentia Labs
About Me
I am CTO and Co-founder at Absentia Labs, a company focused on developing advanced AI solutions for the healthcare industry.
I completed my postdoctoral research at MIT, where I worked on applying generative AI methods to problems in biotechnology. My focus was on integrating machine learning with biological data, ranging from omics datasets to drug discovery pipelines, to develop new computational tools for understanding and engineering biological systems.
Before joining MIT, I earned my Ph.D. in Computer Engineering from the University of Denver, where I specialized in artificial intelligence and natural language processing. During my doctoral studies, I contributed to projects at the intersection of health data, computer vision, and affective computing as part of the computer vision and robotics lab.
My academic foundation was built at Sharif University of Technology, where I focused on natural language processing. This early work sparked my lasting interest in machine learning, software systems, and the mathematical principles underlying complex data and behavior.
Experience
I worked as a senior AI researcher at Speechify, where I focused on developing state-of-the-art speech synthesis and natural language processing systems. My work involved leading research projects, and collaborating with cross-functional teams.
- Developing SOTA speech synthesis based on deep flow-based models
- Developed Phonemizer based on Espeak and deep learning models.
- Developed text Normalizers for speech preprocessing
I was leading a team of AI scientists and software developers to develop a hybrid chatbot for Ryan, a virtual assistant for the company. The project involved using Blender model and AIML technologies to create a chatbot that could understand and respond to user queries in a natural way.
- Develop Hybrid Chatbot for Ryan using Blender model and AIML technologies
- Developed the largest male speech corpus named RyanSpeech in the domain of conversation with high quality for usage in research
- Develop text to speech models based on RyanSpeech to be used in interaction with elderly people who suffer from Dementia and Alzheimer’s disease
- Design and conduct experiments consisting of the dialog system interacting with elderly patients
- Develop an empathetic aware dialog system for Ryan using state of the art models in NLP
- Collaborate with research scientists, development and designing engineers from computer vision, speech, graphics, and game design
I worked as a junior software developer at Batis, where I was involved in developing web applications and software solutions for small businesses. My work included designing user interfaces, implementing business logic, and integrating with various APIs.
- Develop a small scale, custom ERP software in .NET framework following service-oriented architecture for small businesses
- Design user interfaces using WCF for web and desktop application
Education
2023 - 2025
Post-Doctoral in Biomedical engineering
Massachusetts Institute of Technology
Focused on applying generative AI methods to problems in biotechnology, integrating machine learning with biological data, and developing new computational tools for understanding and engineering biological systems.
- Large Language models for scRNA data, Multimodal Learning for Mapping the Genotype-Phenotype Dynamics
- Research on interpretability methods on Genotype-Phenotype relationships
- Research on Diffusion based models for temporal analysis of developmental biological processes
2017 - 2023
PhD in Computer Science
University of Denver
I Publiced two papers in the ACM SIGCHI Conference on Artificial Intelligence.
- Topical language generation using Latent Dirichlet Allocation and GPT transformers. Controlling different aspects of language generation and simulating the style of given documents
- Empathetic dialog system based on the state-of-the-art deep learning transformer language models. Design a novel approach to use multi-task learning for incorporating emotion and other contextual knowledge to improve language generation and response quality.
- Common sense abduction language generation and inference using large language models and temporal reasoning, context filtering, and semantic entailment.
- Augmented goal-oriented conversational agent for interacting with elderly people. Designing an improved AIML-based dialog system that is used within a humanoid robot and integrate non-text clues like vision to be able to interact with people more naturally.
- Developed SOTA text to speech and vocoder models for newly created speech corpus
2014 - 2016
Master of Science in Computer Engineering
Sharif University of Technology
I completed my master’s degree in computer engineering with a focus on natural language processing. My thesis was on developing a novel approach to find temporal event in text.
- Creating a temporal corpus for Persian language and developing a system for recognizing events and temporal relations between them.
Get in Touch
My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!