Research Engineer – Seismology & Full Waveform Inversion (H/F)

Mantle8; Toulouse; On-site; EUR 55,000 - 90,000; Full time

Benefits offered by this job

Health insurance fully covered for the

Meal vouchers

CESU

Job summary

Mantle8 is seeking a seismologist to build an in-house full waveform inversion capability for passive seismic data and production seismic imaging. The role blends R&D and production, with responsibilities spanning workflow design, forward modeling, inversion components, and application to field datasets.

You will work with open-source codes and run inversions on HPC/cloud infrastructure. Desirable skills include Python/C/C++, HPC, and a PhD in seismology or geophysics.

Qualifications

• PhD in seismology, geophysics, applied mathematics or a related field, with a thesis focused on full waveform inversion
• Early career: recent PhD graduate, up to 5 years after PhD
• Hands-on development of FWI code: written, tested and benchmarked significant parts of an inversion workflow – not only used commercial software as a black box
• Complete view of the FWI protocol, from data to model, and in-depth understanding of each component
• Solid background in seismology and wave propagation
• Experience with HPC / parallel computing

Responsibilities

• Design, develop and adapt an FWI workflow for passive seismic data, covering the full chain from raw data to velocity model
• Build and optimize forward-modeling and adjoint-based inversion components: gradient, misfit functions, regularization, uncertainty
• Benchmark on synthetic models, then apply to Mantle8 field datasets
• Assess and extend open-source codes (SPECFEM, Salvus, Devito, SeisFlows, ObsPy…), with attention to licensing
• Run and optimize inversions on HPC / cloud infrastructure
• Process dense nodal-array datasets (Stryde, SmartSolo): pre-processing, Green's-function retrieval, source characterization
• Apply established methods – ambient-noise tomography, traveltime tomography – to deliver models within project timelines
• Share clear results with geologists and geochemists and integrate them with other geophysical data
• Contribute to acquisition design (array geometry, sensor selection)
• Document methods and code, and write technical reports
• Months 1–3: take ownership of data and tools, choose the code base, first synthetic tests
• By month 6: first FWI results on Mantle8 field data

Skills

Python

C/C++

Fortran

Julia

Git

HPC / parallel computing

English

Education

PhD in seismology/geophysics

Thesis focused on FWI

Tools

SPECFEM

Salvus

Devito

SeisFlows

ObsPy

Job description

Job Description

Passive seismic data (ambient noise, microseismicity, teleseismic events) are noisy and irregular, and off-the-shelf tools rarely fit. We are looking for a seismologist who understands full waveform inversion (FWI) from the inside – physics, mathematics, algorithms and code – to build an in-house FWI capability based on code we understand, control and can modify.

The role combines R&D and production.

R&D – Full waveform inversion

• Design, develop and adapt an FWI workflow for passive seismic data, covering the full chain from raw data to velocity model
• Build and optimize forward-modeling and adjoint-based inversion components: gradient, misfit functions, regularization, uncertainty
• Benchmark on synthetic models, then apply to Mantle8 field datasets
• Assess and extend open-source codes (SPECFEM, Salvus, Devito, SeisFlows, ObsPy…), with attention to licensing
• Run and optimize inversions on HPC / cloud infrastructure

Production – Seismic imaging for exploration

• Process dense nodal-array datasets (Stryde, SmartSolo): pre-processing, Green's-function retrieval, source characterization
• Apply established methods – ambient-noise tomography, traveltime tomography – to deliver models within project timelines
• Share clear results with geologists and geochemists and integrate them with other geophysical data
• Contribute to acquisition design (array geometry, sensor selection)
• Document methods and code, and write technical reports

First milestones

• Months 1–3: take ownership of data and tools, choose the code base, first synthetic tests
• By month 6: first FWI results on Mantle8 field data

Requirements

Required

• PhD in seismology, geophysics, applied mathematics or a related field, with a thesis focused on full waveform inversion
• Early career: recent PhD graduate, up to 5 years of experience after PhD
• Hands-on development of FWI code: you have written, tested and benchmarked significant parts of an inversion workflow – not only used commercial software as a black box
• Complete view of the FWI protocol, from data to model, and in-depth understanding of each component
• Solid background in seismology and wave propagation
• Strong scientific programming (Python and/or C/C++, Fortran, Julia), numerical methods and Git
• Experience with HPC / parallel computing

Desirable

• Passive seismology: ambient-noise interferometry, teleseismic or microseismic data
• Ambient-noise or traveltime tomography in an applied setting
• Machine learning for seismic data
• Joint inversion or exploration geophysics

Personal qualities

• Adaptability when working with complex, noisy data
• Autonomy, rigour and the ability to deliver on deadlines
• Clear communication within a multidisciplinary team
• Working proficiency in English; French is a plus

Non-linear career paths are welcome: what matters most to us is your expertise.

What we offer

• A central role in building a new capability in natural hydrogen exploration
• Access to proprietary datasets, with freedom to publish and present at conferences
• An interdisciplinary geoscience team and modern computing resources
• Health insurance fully covered for the employee, meal vouchers, CESU

Recruitment process

• HR interview
• Technical interview with the Geophysics Lead
• Final interview with the CTO