AI Engineer

Company Description

Mentaily is an innovative mental health company leveraging AI and GenAI to enhance access to mental health care. Its flagship product, Liv, is an award-winning, AI-powered Clinical Decision Support System (CDSS) that accelerates psychiatric intake processes, delivering precise and DSM-5-compliant clinical summaries. Liv integrates advanced machine learning, culturally adaptable care plans, and real-time safety assessments to provide personalized and effective mental health solutions. Backed by partnerships with prestigious organizations like Sheba Medical Center and Microsoft, Mentaily is dedicated to transforming mental health care delivery with AI-driven precision, reducing wait times, and improving patient outcomes.

Role Description

Join our core AI team building LIV — a clinical intelligence platform that transforms psychiatric assessments using AI-driven automation. Work at the intersection of machine learning, NLP, and clinical data science to design, build, and optimize AI systems powering real-world mental health evaluations across defense, veterans’ affairs, and public health organizations.

This role blends hands-on ML engineering with analytical rigor — building production AI pipelines, analyzing clinical data patterns, and collaborating closely with clinicians and product teams to deliver AI that is safe, explainable, and clinically grounded.

Qualifications

3+ years experience in applied ML and NLP 

2+ years hands-on with LLMs, prompt engineering, and vector search / embeddings

Experience with RAG frameworks (LangChain, LlamaIndex) and agent orchestration (LangGraph, AutoGen, CrewAI)

Experience building ML pipelines 

Hands-on experience with cloud AI platforms — Azure (preferred), AWS, or GCP

Knowledge of MLOps practices: CI/CD for ML, model monitoring, containerization (Docker), API deployment

Strong Python proficiency and modern ML stack (PyTorch, Hugging Face, scikit-learn, pandas)

Experience with model evaluation for LLM reliability, safety, hallucination detection, and explainability

Strong background in experimental design, prompt evaluation, and data science research methodologies for validating and improving LLM accuracy, consistency, and stability in production

Excellent collaboration and communication skills — ability to translate 

complex ML/data concepts into clear language for clinicians, PMs, and leadership

Responsibilities and authorities:

Design, develop, and optimize NLP/LLM models for clinical classification, diagnostic scoring, and personalized assessment flows

Build and maintain end-to-end ML pipelines — from data ingestion and feature engineering to model training, evaluation, and deployment

Implement and refine RAG architectures, prompt engineering strategies, and agent orchestration for multi-step clinical workflows

Fine-tune LLMs and optimize prompt-response chains for reliability, safety, and clinical accuracy

Manage model lifecycle: versioning, A/B testing, monitoring, drift detection, and continuous improvement

Analyze real-world clinical usage data to identify patterns, improve AI outcomes, and validate diagnostic accuracy against DSM-5 criteria

Design experiments and evaluation frameworks to measure model performance, bias, and safety across patient populations

Build dashboards and data visualizations to communicate clinical AI insights to stakeholders

Apply data science research methodologies to bridge research-to-production gaps — translate experimental results into robust, deployable AI models with measurable clinical reliability benchmarks

Partner with clinical experts, product managers, and engineering teams to ensure ethical, compliant, and safe AI

Required Training and Qualification:

M.Sc. or Ph.D. in Computer Science, Data Science, Statistics, Machine Learning, NLP, or a related quantitative field. B.Sc. with exceptional hands-on experience will also be considered.

Nice-to-Have:

Experience in healthcare AI, clinical NLP, or regulated environments (HIPAA, MDR)

Familiarity with DSM-5 diagnostic frameworks or clinical assessment workflows

Experience with data visualization tools (Chart.js, Plotly, Power BI) and building analytical dashboards

Background in experimental design, A/B testing, and causal inference

Experience with Azure-specific AI services (Azure OpenAI, Cognitive Services, Azure ML)

Proficiency with AI-powered development tools (Claude Code, Cursor, GitHub Copilot, Windsurf) for accelerated prototyping, code