Dyania Health Tech Stack

NLP-powered EMR analysis for clinical trial patient matching

Hospitals and Health Care Jersey City, New Jersey 11–50 employees Founded 2019 Privately Held

Dyania Health builds NLP and transformer models trained on clinical notes to automate patient identification for clinical trials. The tech stack—Python, PyTorch, Hugging Face, DeepSpeed—reflects a research-first approach to biomedical NLP; the hiring mix (equal engineering and research roles, weighted toward senior talent) shows the company is productionizing models rather than just publishing papers. Pain points center on manual EMR review and trial eligibility matching, which the platform targets directly.

Tech Stack 17 technologies

Core StackPython PyTorch Java Kotlin Scala C++ Jira Hugging Face AWS Docker Git DeepSpeed REST API SQL NoSQL GCP Azure

What Dyania Health Is Building

◆Challenges

Manual chart review of emrs
Manual review of electronic medical records
Clinical trial criteria matching
Manual chart review inefficiency
Clinical trial eligibility matching
Data extraction from emrs

▲Active Projects

Ml-powered microservices for biomedical information processing
Productionization of nlp and transformer-based models for biomedical information processing
Extracting medical information from structured and unstructured data sources
Biomedical information processing microservice architecture
Prototype new functionality for biomedical text processing
Retrospective observational research protocols
Evidence-guided care pathways
Real world insights
Develop nlp models for emr data extraction
Prototype new functionality with product and clinical teams

Hiring Activity

Steady8 roles · 3 in 30d

Department

Engineering

Research

Seniority

Senior

Mid

Company intelligence

Find more companies like Dyania Health by tech stack, pain points and active projects

Get started free

About Dyania Health

Dyania Health develops a platform that uses natural language processing to extract and analyze data from electronic medical records, with a primary focus on matching patients to clinical trials. Founded in 2019 and based in Jersey City, the company employs a team of clinicians and NLP applied scientists. The platform processes EMR data in half a second per record, addressing the bottleneck of manual chart review across research and care coordination workflows. The company serves research institutions and healthcare systems conducting or recruiting for clinical trials.

HeadquartersJersey City, New Jersey

Company Size11–50 employees

Founded2019

Hiring MarketsUnited States, Greece

Frequently Asked Questions

What machine learning frameworks does Dyania Health use?

PyTorch and Hugging Face for NLP, plus DeepSpeed for model optimization. Engineering stack also includes Python, Java, Kotlin, Scala, and C++ across AWS, GCP, and Azure.

What is Dyania Health working on?

Productionizing NLP and transformer models for EMR data extraction, building microservices for biomedical information processing, and developing disease-specific models for clinical trial patient matching and care pathway optimization.