TetraScience is the Scientific Data and AI Company building Tetra OS, the operating system for scientific intelligence. We help the world’s leading life sciences firms turn fragmented scientific data into AI-native assets and scientific workflows that accelerate discovery, development, and manufacturing. TetraScience’s growing ecosystem of strategic partners includes NVIDIA, Databricks, Thermo Fisher Scientific, Snowflake, Google, and Microsoft.
In connection with your candidacy, you will be asked to carefully review “The Tetra Way,” authored by our CEO, Patrick Grady; it is impossible to overstate the importance of this document, and you should take it literally as you decide whether our mission, culture, and expectations are right for you.
The RoleTetraScience is the scientific data and AI company. Our documentation is how customers, from bench scientists to platform engineers, learn to build on the platform, and increasingly it is how AI agents consume the platform too. We are looking for a Documentation Engineer to own documentation as a system: the pipelines that build and publish it, the AI-augmented workflows that generate drafts for human review and refinement, the review and publish process, and the infrastructure that makes it reliably consumable by AI agents.
This is primarily a documentation systems role, not only a writer who uses tools. The differentiator is building and owning the systems that produce, validate, publish, and AI-enable our documentation. Strong writing and editorial judgment are still required, but the center of gravity is tooling and systems, and a large portion of the day to day is building.
You will lead, not just maintain. You will take our existing docs-as-code foundation and AI-assisted documentation workflows and grow them into a docs-as-AI-agents capability that is differentiated for a life-sciences AI platform. You will still own editorial quality and the release-notes cadence, but you will spend most of your time building leverage rather than absorbing work.
Own the documentation site and its publishing as software: the docs-as-code repo, the CI/CD publishing pipelines, build performance, and automated link, structure, and quality checks.
Build and grow AI-augmented documentation workflows: AI-assisted drafting, summarization, classification, consistency and staleness checks, and a feedback loop that improves generation quality over time, all with human oversight.
Build our docs-as-AI-agents position: structure and transform content so AI systems can reliably chunk, index, and reason over it, and stand up and maintain MCP-style interfaces so agents and assistants consume our docs accurately.
Generate reference documentation from source (OpenAPI and related specs) and keep docs in lockstep with the platform as code changes.
Lower the barrier for internal contributors (PMs, squad leads, engineers) to ship their own docs through the docs-as-code workflow, and reduce repetitive work through automation.
Own the release-notes and customer-communications cadence that goes out with every platform release, and run the SME review that keeps it accurate and on time.
Own the documentation style guide, hold the review-and-publish gate, and keep the team runbook current so the function is not dependent on any one person.
RequirementsBasics Requirements
- 5+ years owning documentation tooling, content engineering, or developer documentation for a developer-platform or enterprise B2B product.
- Engineering ability in a scripting or web stack (for example Python, TypeScript, or JavaScript) and real fluency with docs-as-code: Git, pull-request review, CI/CD, and a static-site or CMS publishing pipeline.
- Hands-on experience building AI-augmented or LLM-backed workflows: integrating LLM APIs, AI-assisted authoring, and structuring content for AI consumption.
- Ability to read and reason about a real codebase and API surface well enough to document it accurately and to build tooling against it.
- Strong editorial judgment: you can take a dense engineering change and make it clear, correct, and customer-safe.
- Bachelors or Masters degree in a technical field, or equivalent practical experience.
- Experience making documentation consumable by AI agents (llms.txt, content negotiation, RAG pipelines, MCP servers)
- Experience in BioPharma or scientific software, or in regulated and validated (GxP) environments.
- Experience generating reference docs from OpenAPI or related specifications with two-way Git sync.
- Developer-relations or developer-education exposure.
Benefits
- Competitive Salary and equity in a fast-growing company.
- Supportive, team-oriented culture of continuous improvement.
- Generous paid time off (PTO).
- Flexible working arrangements - Remote work.
We are not currently providing visa sponsorship for this position
Skills Required
- 5+ years owning documentation tooling, content engineering, or developer documentation for a developer-platform or enterprise B2B product.
- Engineering ability in a scripting or web stack (for example Python, TypeScript, or JavaScript) and fluency with docs-as-code: Git, pull-request review, CI/CD, and a static-site or CMS publishing pipeline.
- Hands-on experience building AI-augmented or LLM-backed workflows: integrating LLM APIs, AI-assisted authoring, and structuring content for AI consumption.
- Ability to read and reason about a real codebase and API surface well enough to document it accurately and to build tooling against it.
- Strong editorial judgment: convert dense engineering changes into clear, correct, customer-safe documentation.
- Bachelors or Masters degree in a technical field, or equivalent practical experience.
- Experience making documentation consumable by AI agents (llms.txt, content negotiation, RAG pipelines, MCP servers).
- Experience in BioPharma or scientific software, or in regulated and validated (GxP) environments.
- Experience generating reference docs from OpenAPI or related specifications with two-way Git sync.
- Developer-relations or developer-education exposure.
What We Do
Biopharma R&D runs on data trapped in silos. Every instrument, ELN, and LIMS writes results in its own format, so most of what a scientist learns in one experiment never reaches the next. That's a big reason drug development costs keep climbing even as computing power gets cheaper. TetraScience builds Tetra OS, the operating system for scientific intelligence. It sits above and alongside a company's existing instruments, ELNs, and LIMS, works across any vendor, and turns raw scientific data into structured, AI-ready intelligence. Four components make it work. The Scientific Data Foundry ingests data from any source and converts it into governed, machine-readable data at scale. The Scientific Use Case Factory turns one-off workflows into reusable, validated ones deployed across sites. Tetra AI reasons and orchestrates on top of that foundation, grounded in real data instead of guessing. Sciborg teams, our embedded scientist-engineers, work inside customer organizations to get all of it adopted, not just installed. These parts compound. Better data makes workflows faster to build. Workflows generate better data. The result is intelligence that gets more valuable with use instead of resetting on every new project. Customers include Takeda, Bayer, Regeneron, Roche and and AstraZeneca, alongside partners like NVIDIA, Databricks, and Thermo Fisher Scientific. One top-20 pharma customer connected over 1,000 instruments, processed a petabyte of data in a year, and cut 20,000 hours of administrative work. Founded by Patrick Grady and Spin Wang, headquartered in Boston with a European base in Basel. Mission: radically improve and extend human life.
Why Work With Us
We're building the operating system underneath scientific R&D, not another point tool. Work touches real drug development at Takeda, Bayer, and Novartis. Every employee sees the same financials the board sees. We hire people who reason from first principles and run at hard problems.
Gallery
.png)
