Senior Life Sciences Analyst, AI Model Training

Senior Life Sciences Analyst, AI Model Training

Company:
Norstella

Location:

Remote, United States

Date Posted:
Jul 24, 2026

Employment Type:

Full Time

Job : R-2075

Senior Life Sciences Analyst, AI Model Training

Why Norstella? Norstella unites market-leading companies that all have a shared goal of improving patient access. Each organization (Evaluate, Citeline, MMIT, Panalgo, The Dedham Group) delivers must-have answers for critical strategic and commercial decision-making.

• Assess the market need and competitive landscape
• Know precisely which drugs to prioritize in their portfolios
• Find out where the launch difficulties will be—before they’re difficulties
• Track and improve market access post-launch

By combining the efforts of each organization under Norstella, we can offer an even wider breadth of expertise, cutting-edge data solutions and expert advisory services alongside advanced technologies such as real-world data, machine learning-driven predictive analytics. At Norstella, we don’t just deliver information and insights. We deliver answers you can act on.

As a Senior Life Sciences Analyst for Model Training at Norstella, you will sit at the intersection of deep clinical and scientific domain expertise and applied AI development. This role will be embedded within a group of life science thought leaders, but will interface across cross-functional teams of data scientists, machine learning engineers and data engineers. Your work centers on the preference and judgment layer of model development — through scoring and ranking outputs, you will calibrate models on dimensions of quality, accuracy and clinical reasoning such that they are able to handshake with persona use cases.

Your judgment becomes

the signal that bridges a capable model and a genuinely useful one, and will play a critical role in our efforts to deliver predictive analytics and insights across clients.

• Evaluate model outputs across clinical, regulatory, and life sciences tasks by rating individual responses, ranking pairs and lists, and rewriting outputs to demonstrate the preferred response.
• Define and maintain evaluation rubrics, preference dimensions, and quality criteria (factual accuracy, clinical reasoning quality, etc) that govern model outputs and preference alignment.
• Act as an expert-in-the-loop for how life science models are commercialized with clients, including ongoing support for implementation.
• Advise and assist with interpretation of model behavior, bridging the desired behavior back to the dataflow inflection point, working with the relevant teams to adjust as needed.
• Partner with data scientists and MLEs to walk feedback into data collection pipelines.
• Conduct red-team and adversarial evaluation of models, surfacing subtle failure modes — hallucinations, citation errors, unsupported clinical claims, regulatory missteps — that automated metrics cannot reliably catch.
• Analyze preference data and model behavior to identify systematic gaps, then partner with the fine-tuning SMEs and the data science team to decide how each gap is addressed through preference data.
• Conduct new proofs of concept for novel domain capabilities.
• Contribute to Norstella’s knowledge graph and taxonomy work and help design new agentic workflows based on domain-grounded language models.

01:
Bold, Passionate, Mission-First

We have a lofty mission to Smooth Access to Life Saving Therapies and we will get there by being bold and passionate about the mission and our clients. Our clients and the mission in what we are trying to accomplish must be in the forefront of our minds in everything we do.

02:
Integrity, Truth, Reality

We make promises that we can keep, and goals that push us to new heights. Our integrity offers us the opportunity to learn and improve by being honest about what

works and what doesn’t. By being true to the data and producing realistic metrics, we are able to create plans and resources to achieve our goals.

03:
Kindness, Empathy, Grace

We will empathize with everyone's situation, provide positive and…