Senior Machine Learning/Data Engineer

PsychPlus; Pakistan; Remote; PKR 41,528,000 - 52,602,000; Full time

Job summary

PsychPlus is redefining how mental health care is delivered through a digital-first approach, with in-person offices and virtual care. We seek a Senior Machine Learning/Data Engineer to build models across clinical and operational data, deploy them to production, and own their lifecycle.

You will work closely with AI, data platform, engineering, clinical, and revenue-cycle teams. You will span applied ML, production engineering, and data analytics, delivering impactful insights such as

Qualifications

• 6+ years building machine learning models professionally.
• 3+ years deploying and operating models in production.

Responsibilities

• Build, evaluate, and ship ML models to production with feature pipelines and deployment.
• Build monitoring, alerting, and retraining processes to keep models healthy.
• Develop backend services/APIs to serve model predictions in production.
• Understand and shape data from EHR and claims systems; document discoveries.
• Design and productionize metrics and scheduled lakehouse jobs.
• Perform data insights analyses and present findings to technical and non-technical stakeholders.
• Create dashboards for operational, clinical and leadership audiences.
• Design controlled rollouts and A/B tests to measure real impact.
• Contribute to applied research in digital phenotyping and clinical trajectory modelling.

Skills

Machine learning

Python

PySpark

SQL

Backend services

Cloud data platforms

MLOps

Data wrangling

Fluent English

Survival analysis

Tools

Power BI

Databricks

Azure

AWS

GCP

Microsoft Fabric

Job description

The current delivery model for mental health care is broken in this country. PsychPlus set out on a mission to reimagine how mental health care is delivered to folks who need it. Our goal is to provide a digital-first, modern approach to psychiatry and therapy. We provide care both in-person at our offices across Texas and virtually so our patients can be seen at their convenience.

Through a combination of exceptional medical and psychological care and best-in-class technology, we provide an unparalleled approach that serves our patients' needs in an integrated way. Join us in our mission to ensure that every person has access to affordable and accessible mental health care.

About The Role

We are looking for a Senior Machine Learning/Data Engineer to join our Data/AI/ML team. This is a hands-on, end-to-end role; you will build machine learning models, put them into production, and be responsible for them once they are running.

The role deliberately spans three areas that are often separated. You will do applied machine learning (build and evaluate models on real clinical and operational data), production engineering (deploy, serve and monitor them reliably), and data and analytics (understand messy source data, define the metrics the business runs on, and turn analysis into decisions).

You will work directly with the AI Team and closely with our data platform, engineering, clinical and revenue-cycle teams. Because the team is small, your work has immediate visibility; models you build are discussed at executive level within weeks of going live.

Examples of the projects that you will work on include (but not limited to): appointment attendance prediction, model performance and iteration, revenue-cycle analytics, prediction of mental health states, core business metrics, etc.

Responsibilities

• Build, evaluate, and ship models. Develop ML models on clinical, and operational data, and take them all the way to production, including feature pipelines, training, evaluation, deployment and rollout.
• Build the monitoring, alerting and retraining processes that keep models healthy. Detect drift and degradation, and diagnose root causes when a model is not delivering.
• Build and maintain the backend services and APIs through which our EHR consumes model predictions. Write production-grade, tested, version-controlled code — batch and request-response.
• Understand and shape the data: Go deep into our EHR and claims systems to work out what the data actually means, where it is wrong, and how to make it usable. Document what you learn so the next model is faster to build.
• Design and implement the metrics that the AI team as well as the company relies on, with defensible definitions, and productionize them as scheduled jobs in the lakehouse.
• Data insights: run the analyses that answer real operational questions (such as which patients, which providers, which payers, what changed and why) and present findings clearly to technical and non-technical stakeholders.

As a secondary responsibility, you may also contribute to the following:

• Dashboards and reporting: Build and maintain dashboards for operational, clinical and leadership audiences.
• Design and analyze controlled rollouts and A/B tests to measure the real impact of models and product changes.
• Applied research: Contribute to novel work in digital phenotyping, clinical trajectory modelling and behavioral signal processing, alongside a team with a strong peer-reviewed publication record and an active patent portfolio.

Requirements

• Fluent in English, written and spoken. You will work daily with colleagues in Europe and the United States.
• Machine learning experience: 6+ years building machine learning models professionally, including at least 3 years of models you personally deployed and operated in production on live data.
• Strong understanding of gradient boosting and tree-based methods, feature engineering on tabular and temporal data, class imbalance, probability calibration, and choosing operating thresholds under real business constraints. Deep-learning experience is welcome but this role is mostly not that.
• Programming: Expert-level Python for production systems — modular, tested, version-controlled code (not only notebooks!). Advanced PySpark. Expert SQL, including complex joins, window functions and query optimisation over large schemas.
• Backend engineering: Proven experience building and deploying backend services or APIs (REST or GraphQL) that serve predictions in production. You should be comfortable owning the service, not just the model.
• Cloud data platforms: 3+ years working on a modern cloud data platform. Microsoft Fabric or Azure strongly preferred; substantial Databricks, AWS or GCP experience will also be considered.
• MLOps: Hands-on with experiment tracking, model versioning, reproducible training pipelines, and production monitoring (for example MLflow or equivalent).
• Demonstrated experience with “ real-world messy data” such as poorly documented, and inconsistent data — reverse-engineering source-system behaviour, finding silent data-quality failures, and being right about what the data means.
• (Nice-to-have) Healthcare data experience — EHR, claims, billing or clinical records — and work under HIPAA, or comparable data-protection requirements.
• (Nice-to-have) Dashboarding with Power BI or similar, including ownership of the underlying semantic model.
• (Nice-to-have) Experience with survival analysis, uplift modelling, causal inference or optimisation under constraints.

Soft Skills

• Intellectual honesty: you report what the data says, including when it undermines a result you hoped for. You are direct about model limitations and comfortable telling stakeholders that a number is weaker than it looks.
• Communication: you explain technical decisions clearly, with pros and cons, adjusted to the audience. You ask questions to understand context before building.
• Growth mindset: You enjoy debugging a broken pipeline or an unexplained metric as much as building a new model. You consider a bug as an opportunity to improve the system, not an inconvenience.
• Team player: A strong collaborator, willing to work outside your comfort zone, and comfortable to raise the engineering standard around you.