Your mission
Turn Dunia's experimental output into an understanding that drives better decisions
Dunia is building AI for one of the hardest unsolved problems in science: turning materials discovery from an academic, trial-and-error process into a programmable, scalable discipline.
We run high-throughput experiments, simulations, and ML-driven optimization loops. The hardest problem is not generating data, and as our lab throughput increases, so does our data output. It is knowing what that data actually means.
As Chemical Data Scientist, you will sit at the center of Dunia’s discovery loop. You will be the person who spots the signal from the noise, and whether a trend is real or misleading. You will ensure that the organization, its systems and models are actually learning.
This role is about judgment as much as analysis. You will shape how experimental evidence flows into models, simulations, and decisions, and in doing so, how fast and how well Dunia discovers new materials.
Dunia is building AI for one of the hardest unsolved problems in science: turning materials discovery from an academic, trial-and-error process into a programmable, scalable discipline.
We run high-throughput experiments, simulations, and ML-driven optimization loops. The hardest problem is not generating data, and as our lab throughput increases, so does our data output. It is knowing what that data actually means.
As Chemical Data Scientist, you will sit at the center of Dunia’s discovery loop. You will be the person who spots the signal from the noise, and whether a trend is real or misleading. You will ensure that the organization, its systems and models are actually learning.
This role is about judgment as much as analysis. You will shape how experimental evidence flows into models, simulations, and decisions, and in doing so, how fast and how well Dunia discovers new materials.
Your tasks will include:
Be the scientific sense-maker- Interrogate data from ongoing electrocatalyst campaigns
- Identify patterns, anomalies, and failure modes that others miss
- Develop intuition for where experiments lie, and where they tell the truth
- Decide which experimental signals should inform ML feature design
- Decide which experimental signals and discrepancies are important enough to merit computational explanation.
- Give lab and automation teams concrete feedback on experimental quality and design
- Create clear, concise digests that align the entire team on what was learned
- Track how understanding evolves across campaigns, not just within them
- Raise the bar for how scientific progress is communicated internally
- Shape data pipelines and analytical views by using them aggressively
- Help define what “analysis-ready” data actually meansin practice
- Ensure infrastructure evolves aroundreal scientific workflows, not abstractions