.png){kind=logo}
Ramiz Qudsi
Principal Data Scientist & Software Engineer specializing in space mission data systems
Connect with Ramiz
Ramiz already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
Space/heliophysics ML engineer who built a PyTorch GRU model to propagate solar wind from L1 to the magnetopause with probabilistic outputs for uncertainty quantification, achieving ~25% better CRPS than standard approaches. Also developed production-grade Python ETL and an open-source telemetry processing package for a mission (LEXI), using Docker and GitHub Actions CI/CD and iterating with scientist/engineer stakeholders.
Experience
Education
Key Strengths
- Built GRU-based probabilistic solar wind propagation model with uncertainty quantification (mean/variance outputs for 9 parameters)
- Improved model performance ~25% (average CRPS) versus typical field model via structured fine-tuning and Hyperband search
- Strong time-series ML rigor: designed data splits to mitigate temporal leakage in highly autocorrelated signals
- Validated data cleaning and model outputs using multi-spacecraft cross-checks (Wind and MMS/MMS1 in-situ observations)
- Production-minded Python engineering: delivered ETL pipeline + open-source package with Docker and GitHub Actions CI/CD
- Pragmatic delivery and prioritization: shipped core Level-1 processing/visualization under mission timelines and iterated via PRs/releases
Discover more candidates like Ramiz
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent