Research Software Engineer
We are looking for a Research Software Engineer to help neuroscience labs convert their data to Neurodata Without Borders (NWB) and publish it on the DANDI Archive. Each project starts with a lab's own acquisition systems, file formats, and analysis outputs, and ends with a standardized, well-documented dataset that others can reuse. The work is roughly evenly split between building conversion pipelines for specific labs and improving the open-source tools those pipelines are built on, including NeuroConv, PyNWB, and the NWB Inspector. We rely heavily on agentic LLM-based development in this work, using coding agents such as Claude Code to read unfamiliar formats, draft conversion code, and run tests, with engineers directing the agents and reviewing what they produce. You would join a small, fully remote team that works closely with each other and with the research groups we support. This is a contract position paid hourly, at $60 to $75 per hour depending on experience. We are open to anything from 25 hours per week up to full time. For candidates based in the United States, we are also open to a full-time salaried position with benefits, at $95,000 to $120,000 per year depending on experience.
Responsibilities
- Build and maintain NWB conversion pipelines for electrophysiology, optical physiology, behavior, and related experimental data
- Meet with lab members to understand their experiments, data formats, and metadata, and translate that into accurate NWB files
- Prepare datasets for publication on the DANDI Archive and help labs through the upload and validation process
- Contribute features, bug fixes, and documentation to open-source projects such as NeuroConv, PyNWB, HDMF, and the NWB Inspector
- Use coding agents to speed up conversion and tool development, and help improve the workflows, prompts, and tooling our team uses to work with them
- Review code from teammates and outside contributors, and write tests for the code you contribute
- Manage several projects at once and keep labs informed of progress and open questions
Requirements
- Strong Python skills, including experience with NumPy and working with large scientific datasets
- Experience with a scientific data format such as HDF5, Zarr, or a vendor format used in neuroscience
- A background in neuroscience, biomedical engineering, or a related field, or substantial experience working with neuroscience data
- Hands-on experience developing software with LLM-based coding agents (for example Claude Code, Codex, or Cursor), and the judgment to check their output carefully rather than accept it as written
- Familiarity with Git, GitHub, and collaborative open-source development
- Clear written and spoken communication in English, and comfort working directly with researchers
- Ability to work independently in a remote team and overlap several hours a day with European working hours
- Experience with NWB, DANDI, SpikeInterface, or related tools is a plus but not required
How to Apply
Please email info@catalystneuro.com with the subject line “Application: Research Software Engineer” (the button below fills this in for you), and include:
- Your CV
- A link to your GitHub profile or another code sample
- A short note, a few paragraphs at most, answering two questions:
- Describe a dataset you have worked with and what made it difficult to handle.
- Describe how you have used coding agents in your work, including a time an agent got something wrong and how you caught it.
We will begin reviewing applications on November 10, 2026, and will continue until the position is filled.