ML · Data Eng2025
SynMedix AI
A distributed pipeline that processes roughly 10 GB of electronic health records at 3× the throughput of the serial baseline, with a generative layer on top producing 50,000+ synthetic patient records deployed on AWS SageMaker.
- Throughput over serial baseline
- 3×
- Synthetic records generated
- 50K+
- EHR corpus processed
- ~10 GB
- Real records reproduced
- 0
The problem
Electronic health records are abundant and hard to use at the same time. The volume is large, the privacy constraints are absolute, and serial processing is slow enough that iterating on a model means waiting overnight. A team that wants to train on real clinical data runs into all three at once, and the slow loop is usually what stalls the project.
The approach
Throughput came first. Ingest was restructured around parallel execution with optimised ETL stages, so the corpus can be reprocessed in an afternoon instead of overnight. Privacy came second. A generative model trained on the cleaned corpus learns the joint distribution of the records instead of memorising individual ones, so downstream teams get a dataset they can experiment on freely.
The healthcare industry generates enormous amounts of EHR data but processing it at scale remains painfully slow. I built SynMedix to demonstrate that with the right distributed architecture, medical AI teams can iterate 3x faster on realistic, privacy-safe datasets.
Architecture
- 01
Ingest
Roughly 10 GB of raw EHR pulled into S3, partitioned so downstream stages can be worked in parallel rather than streamed serially.
- 02
Parallel ETL
Optimised transform stages executing concurrently. This is the change responsible for the 3× throughput improvement over the serial baseline.
- 03
Generative layer
PyTorch/TensorFlow models trained on the cleaned corpus to learn the joint distribution across clinical fields.
- 04
Serving
Deployed on AWS SageMaker for scalable inference, with DynamoDB and S3 backing metadata and generated artefacts.
Stack
- Modelling
- PyTorch · TensorFlow · Keras
- Pipeline
- Python · SQL · parallel ETL
- Cloud
- AWS SageMaker · S3 · DynamoDB
- Python
- SQL
- PyTorch
- TensorFlow
- Keras
- ETL
- AWS SageMaker
- AWS S3
- DynamoDB
Outcome
SynMedix emits synthetic patient records that preserve the statistical structure of the source corpus without carrying any individual through it. Downstream teams get a dataset they can train on and share freely. The throughput gain compounds too: experiments that used to cost a night now cost an afternoon.
Live demoRuns in your browser
Explore the latent space.
The generator described above, trained and exported to run in the browser. Move a latent dimension or change the conditioning and the record regenerates locally, with no call to a server.
Generator loads as you reach it…