Machine Learning Engineer - Predictive Modeling, Causal Inference & Interpretability - Sigma Te[...]

Biorce; Austin (TX); Hybrid; USD 140,000 - 210,000; Full time

Benefits offered by this job

Hybrid work model

Private health coverage

MacBook provided

Pet-friendly office

Company events

Job summary

Biorce is seeking a Machine Learning Engineer for the Sigma Team in Austin to own causal inference and predictive modeling for our AI engine predicting clinical trial success. You will lead end-to-end development from research to production, specializing in interpretable, decision-oriented outputs built on uncertain data.

You will collaborate with a Product Owner, Designer, and embedded Scientific Lead to ship a benchmarked model within weeks, applying interpretability techniques and robust

Qualifications

• 5+ years of experience in Machine Learning, Deep Learning, or AI Research.
• Proven proficiency in Python and ML/Deep Learning frameworks (TensorFlow, PyTorch, etc.).
• Solid, hands-on experience with causal inference methods.
• Experience modeling and predicting outcomes from uncertain, noisy, or incomplete data, and translating probabilistic predictions into expected-value or decision-oriented outputs.
• Experience working with data teams to extract structured signals from unstructured or messy data sources for use in modeling pipelines.

Responsibilities

• Own the Model Lifecycle: Lead the full lifecycle of projects leveraging deep learning and causal inference methods, from research to prototyping to final product and monitoring in production.
• Build Inside the Sigma Squad: Work embedded with a Product Owner, a Designer, and the Scientific Lead from day one, taking causal and predictive models from an ambiguous problem to a demoable, benchmarked product in weeks rather than quarters.
• Partner on Data Extraction: Partner with the data team to source, assemble, and extract usable signals from unstructured and semi-structured data for downstream modeling.
• Design Causal Inference Models: Design causal inference models, from classical to deep learning-based methods, to answer counterfactual questions and distinguish what actually drives outcomes from what merely correlates with them.
• Model Outcomes Under Uncertainty: Build models that predict the likelihood of event success under uncertain, noisy, or incomplete data, translating predictions into expected-value estimates to support decision-making.

Skills

Python

Machine Learning

Deep Learning

Causal Inference

Communication

Education

Master’s or Ph.D. in Computer Science, Statistics, or related technical field

Tools

TensorFlow

PyTorch

Vertex AI

Google Cloud Platform

Job description

Biorce is a pioneering Healthtech company dedicated to revolutionizing drug development through the power of AI. We are passionate about accelerating medical advancements and improving patient outcomes.

Our team comprises seasoned clinical research professionals, data scientists, and AI experts, working collaboratively to bridge the gap between cutting-edge technology and real-world clinical needs.

With an unwavering commitment to revolutionize healthcare, we envision a world where all patients benefit from accelerated and cost-effective access to treatments. Biorce is poised to redefine the landscape of healthcare, shaping a future where innovation and accessibility converge for the betterment of humanity.

About the company

Biorce is a pioneering Healthtech company dedicated to revolutionizing drug development through the power of AI. We are passionate about accelerating medical advancements and improving patient outcomes.

Our team comprises seasoned clinical research professionals, data scientists, and AI experts, working collaboratively to bridge the gap between cutting-edge technology and real-world clinical needs.

With an unwavering commitment to revolutionize healthcare, we envision a world where all patients benefit from accelerated and cost-effective access to treatments. Biorce is poised to redefine the landscape of healthcare, shaping a future where innovation and accessibility converge for the betterment of humanity.

About the role

We're looking for a Machine Learning Engineer to join the Sigma Team in Austin and own the causal inference and predictive modeling core of one of our most strategically critical products: an AI engine that predicts the likelihood of clinical trial success, turns that prediction into a decision-oriented, expected-value estimate, and tells sponsors what is actually driving the outcome rather than what merely correlates with it.

You’ll lead the full lifecycle of this work from research through prototyping to a monitored production model, designing causal inference approaches (classical through deep learning-based) to answer counterfactual questions, building predictive models that hold up under uncertain, noisy, or incomplete data, and applying interpretability techniques so the people betting multi-million-dollar development decisions on your model’s output can trust it. You’ll work embedded in a small, cross-functional Sigma squad alongside a Product Owner, a Product Designer, and an embedded Scientific Lead, moving from an ambiguous problem to a demoable, benchmarked model in weeks rather than quarters.

The Sigma Team

Sigma sits inside Biorce’s CSO’s Office. It’s Biorce’s Tech Special Forces, an incubator tasked with building the most disruptive, highest-stakes AI products in the clinical trial space.

This is not a team that iterates on existing products. Sigma builds from zero: defining the strategy, shipping under real speed and ambiguity, and treating every initiative as a new venture. Squads are small and cross-functional, a Product Owner, AI/ML Engineers, an embedded Scientific Lead, and a Designer, all in the room from day one, not brought in once the problem is already framed.

Who We're Looking For

You’re a Machine Learning or AI Research Engineer with a genuine grounding in causal inference, not just predictive modeling, you can articulate the difference between “what predicts the outcome” and “what causes it,” and you design accordingly. You’re comfortable turning probabilistic, uncertainty-laden predictions into expected-value estimates that a non-technical stakeholder can act on, and you don’t treat interpretability as an afterthought, you build it in because the people relying on your models will immediately distrust anything that feels like a black box.

You’re a relentless self-teacher who has picked up new tools, methods, and domains without waiting for a course or a manager to hand you one. You’re comfortable partnering with data teams to wring usable signal out of messy, unstructured data, and with scientists and researchers as collaborators rather than stakeholders. You measure cycle time in weeks, communicate complex AI concepts to diverse audiences, and can lead a technical project autonomously from research question to production model.

Key Responsibilities

• Own the Model Lifecycle: Lead the full lifecycle of projects leveraging deep learning and causal inference methods, from research to prototyping to final product and monitoring in production.
• Build Inside the Sigma Squad: Work embedded with a Product Owner, a Designer, and the Scientific Lead from day one, taking causal and predictive models from an ambiguous problem to a demoable, benchmarked product in weeks rather than quarters.
• Partner on Data Extraction: Partner with the data team to source, assemble, and extract usable signals from unstructured and semi-structured data for downstream modeling.
• Design Causal Inference Models: Design causal inference models, from classical to deep learning-based methods, to answer counterfactual questions and distinguish what actually drives outcomes from what merely correlates with them.
• Model Outcomes Under Uncertainty: Build models that predict the likelihood of event success under uncertain, noisy, or incomplete data, translating predictions into expected-value estimates to support decision-making.
• Quantify Confidence: Use simulation-based or uncertainty quantification methods to gauge confidence in predictions and support robustness checks.
• Write Production-Grade Code: Develop and maintain high-quality, scalable, and robust code in Python.
• Apply Interpretability Techniques: Apply model interpretability techniques to validate model behavior and build stakeholder trust in predictions.
• Run Rigorous Experimentation: Conduct rigorous experimentation and model evaluation, including causal validation and A/B or quasi-experimental testing, to ensure performance and trustworthiness.
• Ship With Product & Engineering: Collaborate with product and software engineering teams to integrate causal and predictive modeling solutions into our core products.
• Stay Current: Stay current with the latest advancements in causal inference, deep learning, and applied statistics research.

Must-haves

• 5+ years of experience in Machine Learning, Deep Learning, or AI Research.
• Proven proficiency in Python and ML/Deep Learning frameworks (TensorFlow, PyTorch, etc.).
• Solid, hands-on experience with causal inference methods.
• Experience modeling and predicting outcomes from uncertain, noisy, or incomplete data, and translating probabilistic predictions into expected-value or decision-oriented outputs.
• Experience working with data teams to extract structured signals from unstructured or messy data sources for use in modeling pipelines.
• Working knowledge of model interpretability and explainability techniques.
• Solid experience architecting, developing, and deploying machine learning models, with a strong foundation in data structures, algorithms, and software engineering principles.
• Ability to work autonomously, lead technical projects, and communicate complex AI concepts to diverse audiences.
• This role is based in our Austin office on a hybrid model, you’ll need to be based in Austin or willing to relocate.

Nice-to-Haves

• Practical experience with deep causal inference approaches, Bayesian modeling, probabilistic programming, or simulation-based methods for uncertainty-aware prediction.
• Familiarity with advanced or mechanistic interpretability methods, beyond standard attribution techniques.
• Exposure to large language models and agentic AI systems, particularly for extracting structured signals from unstructured data or supporting causal/predictive workflows.
• Master’s or Ph.D. in Computer Science, Statistics, or a related technical field with a focus on causal inference or probabilistic modeling.
• Contributions to research (publications in top-tier conferences) or open-source projects in causal inference, probabilistic modeling, or interpretability.
• Experience with Google Cloud Platform and Vertex AI, or open-source causal ML and interpretability libraries.
• Exposure to clinical research, life sciences, or another regulated domain.
• Additional languages, we’re an international team.

Why Join Us?

• A dynamic work environment with an international team, where collaboration and diversity thrive.
• Work alongside top talent, a Product Owner, Designer, and an embedded Scientific Lead — united by a shared purpose and committed to making a real impact.
• Build causal and predictive models that directly inform multi-million-dollar clinical development decisions, on the Sigma Team’s zero-to-one, weeks-not-quarters operating model.
• Direct visibility into the CSO’s Office and the product decisions shaping Biorce’s clinical research portfolio.
• Comprehensive private health coverage to ensure your physical and mental well-being.
• Hybrid work model offering flexibility to balance your professional and personal life.
• Company events to celebrate achievements and enjoy time together.
• Get equipped with a MacBook to enhance your productivity and work experience.
• Our office is pet-friendly! You’ll likely be greeted by a few wagging tails upon arrival.

--