Senior AI Engineer

Talkspace Corporate is hiring a remote Senior AI Engineer to lead the creation of the autonomous AI agents powering our behavioral health platform. If you're driven to put quality mental healthcare within reach of millions, this is your chance to shape technology that genuinely changes lives.

About the Role

As a Senior AI Engineer at Talkspace Corporate, you will serve as a hands-on technical leader, owning the full lifecycle of our multi-agent systems—from cognitive architecture and interaction design through resilient production deployment and continuous learning. The work draws on Large Language Models for reasoning, planning, and goal setting, alongside Reinforcement Learning methods used to safely and ethically model human behavior. Because behavioral health is deeply personal, this remote role places extraordinary weight on safety, transparent autonomy, ethical decision-making, and data privacy. This position is fully remote and operates within Eastern Standard Time; Talkspace's headquarters are located in New York City.

Key Responsibilities

• Architect and build the core agents behind Sentia, including planning modules, memory and retrieval layers, goal-setting logic, and tool-use orchestration.
• Apply Reinforcement Learning, Inverse RL, and control-theory techniques to craft adaptive, long-horizon intervention strategies that lift user outcomes while reducing risk.
• Select, fine-tune, and ship foundation models that drive empathetic, context-aware, and accurate agent communication.
• Model human-agent interaction using insights from cognitive science and behavioral economics so agents stay trustworthy and aligned with therapeutic protocols.
• Build simulation environments to pre-train and stress-test agent policies across diverse psychological scenarios.
• Tune deployment for low-latency, high-resilience multi-agent orchestration.
• Evaluate and champion emerging agentic frameworks and system patterns that keep Sentia ahead of the field.
• Design end-to-end evaluation and observability pipelines—trace-level analysis, offline scoring against golden datasets, live drift monitoring, guardrails, and human-in-the-loop review—plus metrics for reasoning quality, task completion, and safety compliance.

Qualifications

• Bachelor's or Master's in Computer Science, Artificial Intelligence, Robotics, Electrical Engineering, or another rigorous quantitative discipline.
• 5+ years in production software engineering, with 3+ years dedicated to building and deploying complex ML or autonomous systems.
• Advanced Python skills, including its scientific ecosystem.
• Proven MLOps experience: CI/CD pipelines for ML models within service-oriented architectures.
• Practical command of agentic AI—Reinforcement Learning, planning algorithms, and decision-making systems.
• Hands-on LLM work in production: fine-tuning, deployment, prompt engineering, RAG, and cost optimization.
• A track record of clean, scalable, well-tested code, with proficiency in AWS or GCP and Docker/Kubernetes.

About Talkspace Corporate

Talkspace Corporate is on a mission to make therapy accessible to everyone, having already supported more than two million people in receiving quality mental healthcare. We cultivate a diverse, equitable, and inclusive culture where every team member can thrive while advancing the future of mental health.

How to Apply

To apply, complete your application directly on this page, or you'll be redirected to the employer's application platform to finish submitting there.