Lead Commodity Modeling Engineer

Atmospheric AI; Greater London; On-site; GBP 90,000 - 130,000; Full time

Benefits offered by this job

Equity stake

£300/month coaching budget

Remote-first with London hub

Full EOR employment outside the UK

Job summary

Atmospheric AI in the United Kingdom seeks an experienced quantitative engineer to build datasets, models and services turning Brent/naphtha economics into structured signals. You will craft end-to-end pipelines, ship models as production services, and justify outputs to procurement and chemical experts.

The role combines dataset ownership, graph-based price cascades and uncertainty propagation, with a focus on transparent, explainable results and scalable tooling.

Qualifications

• Experience modelling petrochemical or energy price formation, professionally.
• Quantitative modeling under uncertainty with regression, time-series, and disruption analysis.
• Strong Python with the analytical stack and production-grade coding.

Responsibilities

• Model commodity price cascades and regional benchmarks across datasets.
• Forecast and stress-test under disruption with time-series and distributional thinking.
• Build pricing datasets with provenance and lineage; own end-to-end data.
• Ship scenario and sensitivity engines with explicit assumptions and confidence bands.
• Design repeatable systems that ship models rather than one-off analyses.
• Explain methodologies to customers and avoid black-box solutions.
• Collaborate with procurement/chemical experts to align on outputs.

Skills

Python

NumPy

Pandas

NetworkX

Statsmodels

Scikit-learn

PyTorch

ETL/API

Production code

Time-series

Uncertainty modeling

Graph analytics

Tools

Neon Postgres

Dagster

AWS

FastAPI

LangGraph

Pydantic AI

SQL/Data tooling

Job description

Atmospheric AI in the United Kingdom seeks an experienced quantitative engineer to build datasets, models and services turning Brent/naphtha economics into structured signals. You will craft end-to-end pipelines, ship models as production services, and justify outputs to procurement and chemical experts.

The role combines dataset ownership, graph-based price cascades and uncertainty propagation, with a focus on transparent, explainable results and scalable tooling.