Model, harmonize, curate, and validate disease, phenotype, and biomarker data for pharmaceutical clients. Author LinkML schemas, map data to biomedical ontologies, manage controlled vocabularies, review AI-generated curation, define quality checks, integrate disease models with omics domains, and support engineering pipelines. The role also involves scientific stakeholder engagement, client communication, and creating protocols for consistent data curation.
About Rancho BioSciences, LLC
Rancho BioSciences is a fully remote provider of biomedical data curation and data science services for pharma and biotech, spanning drug discovery through translational research. Our teams of scientists, data engineers, and software experts deliver end-to-end solutions across data curation, management, mining, and analysis to help customers accelerate R&D. We partner long-term with blue-chip clients and emerging biotechs around the world, bringing scientific rigor, quality, and a customer-first mindset to every engagement.
About the role
- As part of the larger Service Delivery team, this role will contribute to modeling, harmonizing, and curating disease and biomarker data on behalf of our pharmaceutical clients. You will help scale the production of harmonized, AI-ready data products that include disease phenotypes and biomarkers, working within an agent-assisted harmonization framework. Within the framework, AI does much of the mechanical work and this role would contribute scientific judgment and modeling expertise drive the result.
- The successful candidate combines deep clinical and biomarker domain knowledge, strong information modeling skills, and a genuine affinity for working with AI tools. This is a client-facing role and is expected to partner with both internal (Rancho) and external (client) stakeholders on a regular basis.
- This role is fully remote, but we are looking for candidates in the UK or Western Europe to be client-facing.
What you'll do
- Extend and maintain disease data models covering clinical manifestations, phenotypes, and biomarkers, expanding into new therapeutic areas.
- Harmonize internal and external reference data sources, mapping them to biomedical ontologies and controlled vocabularies; making defensible, documented decisions on ontology reuse, extension, and mapping.
- Utilize and support the evolution of an agent-assisted harmonization framework: direct AI agents through curation and mapping tasks, review and correct their output, and improve the prompts, templates, and QC that guide them.
- Author and version data model artifacts in LinkML, using GitHub for source control.
- Register controlled vocabularies in enterprise terminology systems, following and refining established processes.
- Define validation rules and quality checks for harmonized data, covering ontology term validity, required fields, cross-field consistency, and provenance completeness.
- Integrate disease models with adjacent data domains (e.g., omics schemas) and support engineering teams in building robust pipelines from them.
- Engage scientific SMEs across disease areas to test and review model updates and ensure alignment with intended use.
- Write guidelines and protocols that enable other teams to curate future studies consistently.
- Report progress and decisions clearly to client stakeholders and sponsors.
Qualifications
Required
- PhD in Life Sciences, or MS with several years of hands-on experience with biomarker or clinical data.
- Strong domain knowledge of biomarkers and the clinical manifestations of disease, including how they are measured, reported, and used in translational and clinical research.
- Proven data and information modeling experience for complex biomedical domains, e.g., ER or class diagrams; hands-on experience creating conceptual and logical data models
- Practical experience with LinkML or an equivalent schema framework: classes, slots, ranges, identifiers, constraints, cardinality, and ontology bindings; comfortable authoring YAML schemas.
- Experience with biomedical ontologies and controlled vocabularies covering diseases, phenotypes, anatomy, cell types, assays, and units, and with mapping data to them.
- Solid understanding of FAIR principles, including persistent identifiers, metadata standards, provenance, and versioning.
- Significant hands-on use of AI tools (e.g., LLM assistants, agentic coding tools such as Claude) in scientific or data work.
- Ability to critically review AI output: spot errors, unsupported claims, and wrong ontology mappings, and correct them with sound scientific reasoning.
- Comfortable working with command-line tools.
- Working proficiency in Python or SQL, sufficient to inspect data, run validation, and review AI-generated code.
- Experience working with Git/GitHub.
Nice to Have
- Experience building or operating AI-agent workflows for curation, mapping, or data harmonization.
- Experience with enterprise ontology management or vocabulary systems.
- Experience with omics data.
- Experience with Google Cloud Platform; experience with other cloud data platforms (e.g., Snowflake, Databricks) is a bonus.
Similar Jobs
Cloud • Computer Vision • Information Technology • Sales • Security • Cybersecurity
Leads research and production development of generative AI and large language model systems for cybersecurity. Responsibilities include defining AI research strategy, developing novel reasoning and summarization approaches, selecting model architectures and deployment strategies, building large-scale production systems, mentoring scientists, establishing evaluation standards, and guiding cross-functional technical initiatives. The role also represents the company through publications, speaking engagements, and technical thought leadership.
Top Skills:
Cloud TechnologiesDeep LearningGpu TechnologiesLarge Language Models (Llms)Python
Cloud • Information Technology • Internet of Things • Machine Learning • Software • Cybersecurity • Infrastructure as a Service (IaaS)
Contribute to post-training large language models for telecommunications use cases through fine-tuning, alignment, dataset curation, synthetic data generation, experiment tracking, and GPU-based training. The role involves SFT, parameter-efficient methods such as LoRA and QLoRA, reward training, evaluation, and investigating model performance with experienced engineers and domain specialists.
Top Skills:
Adapter-Based Fine-TuningDpoGpu ComputingGrpoLarge Language Models (Llms)LoraPpoPythonQloraReinforcement LearningSapoSupervised Fine-Tuning (Sft)Synthetic Data Generation
Big Data • Information Technology • Software • Analytics • Energy
Manage and optimize marketing technology platforms, integrations, automations, data governance, and account-based marketing workflows. Responsibilities include maintaining data quality, collaborating with vendors and cross-functional teams, supporting marketing users, documenting processes, delivering platform training, developing dashboards and metrics, and applying AI-enabled workflows. The role requires advanced Martech administration, Salesforce expertise, ABM experience, marketing data analysis, and process transformation skills.
Top Skills:
6SenseAgentforceCloudingoCognismDemandbaseLeandataMicrosoft CopilotMicrosoft Power BiPardotSalesforce Account EngagementSalesforce Crm AnalyticsSalesforce Data CloudSalesforce FlowSalesforce Marketing Cloud NextSalesforce Marketing IntelligenceSalesforce Sales CloudSQLZoominfo
What you need to know about the Manchester Tech Scene
Home to a £5 billion digital ecosystem, including MediaCity, which consists of major players like the BBC, ITV and Ericsson, Manchester is one of the U.K.'s top digital tech hubs, at the forefront of advancements in film, television and emerging sectors like as e-sports, while also fostering a community of professionals dedicated to pushing creative and technological boundaries.
