Research Informatics Software Engineer

Deerfield Management Companies•New York, NY

4d•Hybrid

About The Position

Excelsior Sciences is reinventing small-molecule discovery and manufacturing through Blocc chemistry—modular, automation-friendly chemistry designed for machines to execute and AI to learn from—combined with closed-loop AI learning systems. Backed by a $70M Series A from Deerfield, Khosla Ventures, and Sofinnova, along with a $25M Empire State Development grant, Excelsior is building a lean, high-leverage organization at the intersection of chemistry, automation, software, and AI. Additional investors include Eliヴィトン, Cornucopian Capital, Illinois Ventures, and MIT. Based at the Cure building in New York City, our goal is to build a chemistry and AI-native discovery platform in which high-quality experimental data continuously feeds learning systems that help determine what to make and test next—accelerating the cycle of molecular design, experimentation, and discovery. We are seeking a strong MS Computer Science candidate to join our Research Informatics / R&D IT team. You will help build the digital foundation that enables both scientists and autonomous AI agents—combining cloud-native data platforms, vendor LIMS/ELN and analytical systems, robust scientific data pipelines, and agentic AI / LLM capabilities. This role is ideal for someone with hands-on experience in cloud, data engineering, RAG/multi-agent systems, high-performance ML, and scientific computing who wants to apply those skills at the intersection of lab informatics and AI-native drug discovery. We leverage AI-agile software development and engineering practices as we aim to lead the field in advancing molecule discovery efficiently. What success looks like: Scientific data becomes reliably FAIR and machine-actionable, enabling both human researchers and AI agents to drive faster closed-loop experimentation and accelerate molecule discovery.

Requirements

Master’s degree in Computer Science (or a closely related field) with relevant coursework in cloud computing and the fundamentals of AI and ML.
Demonstrated experience building data pipelines, feature engineering, or scientific data workflows (e.g., Spark/Databricks-style pipelines, data quality checks, performance tuning).
Hands-on experience with cloud platforms (AWS), containers (Docker/Kubernetes), and modern data/backend tools (SQL, PostgreSQL, orchestration frameworks).
Strong proficiency with AI coding assistants and coding agents (e.g., Cursor, Claude Code, GitHub Copilot, or similar tools).
Familiarity with LLM concepts, RAG, retrieval, or multi-agent systems (coursework, projects, or professional exposure).
Willingness and aptitude to rapidly learn commercial LIMS/ELN or analytical platforms (e.g., Genedata, CDD Vault, Virscidian Analytical Studio); prior exposure is a plus.
Proficiency in Python and SQL; additional experience with C++/C, high-performance ML tooling, or scientific computing libraries is a plus.
Strong collaboration skills and ability to work at the intersection of software engineering, data, and scientific applications.

Nice To Haves

Practical, hands-on experience with LLM / agentic AI systems, including RAG, GraphRAG, multi-agent architectures, or production retrieval-augmented pipelines.
Experience optimizing high-performance ML or scientific models (e.g., protein structure prediction, surrogate modeling, Bayesian optimization).
Hands-on work with multi-agent systems, knowledge graphs (Neo4j), or agent frameworks/SDKs.
Familiarity with Airflow or Prefect, PyTorch, and related ML/LLM tooling.
Direct experience with commercial LIMS/ELN or analytical platforms such as Genedata, CDD Vault, Virscidian Analytical Studio, or similar—especially their data models, APIs, and integration points.
Experience with CI/CD, testing, schema design, and production-grade software practices.
Interest in applying agentic AI and robust data engineering to laboratory and drug-discovery workflows.

Responsibilities

Integrate, extend, and support vendor Laboratory Information Management Systems (LIMS), Electronic Lab Notebooks (ELN), and analytical informatics platforms. Scope includes platforms such as Genedata, CDD Vault, Virscidian Analytical Studio, and similar systems—focusing on data models, workflows, APIs, sample/analytical data flows, and connections to instruments and enterprise systems.
Design, implement, and maintain scalable data pipelines and APIs that make scientific data (samples, assays, analytical results, automation streams) FAIR, high-quality, and machine-actionable for both human scientists and AI agents. Leverage modern data platforms, warehouses/lakes, and orchestration tools.
Build and operate cloud-native components (primarily AWS) using containers (Docker/Kubernetes), infrastructure patterns, CI/CD, and workflow orchestration to support lab informatics and AI workloads.
Prototype and productionize agentic AI / GenAI solutions—LLM agents, RAG and GraphRAG systems, multi-agent workflows, and prompt-engineered / retrieval-augmented pipelines—that automate or augment laboratory informatics processes, data interpretation, and closed-loop experimentation.
Collaborate with research scientists and cross-functional engineering teams to translate scientific needs into reliable software, data products, and AI capabilities; contribute to documentation, testing, and knowledge transfer.
Apply software engineering best practices (agile / AI-agile delivery, testing, schema design, performance tuning) in a scientific computing context.
Support continuous improvement of lab digital systems, including data quality, observability, and readiness for AI agents.