• Accelerating adoption of AI within the marketing team by building, managing, and optimizing AI-powered tools and workflows • Deliver a steady cadence of milestones from marketing's AI-first backlog • Help define and track role-based AI utilization goals across the marketing organization • Manage the development and maintenance of MCP connectors across different marketing applications • Set up centralized AI-based “skills” and governance frameworks for marketing AI usage • Lead enablement efforts and host daily office hours to support the marketing team • Share best practices across the team on Skills, Prompts, and Agentic workflows • Design, build, and iterate on agentic AI workflows that automate and accelerate marketing processes • Explore and implement Retrieval-Augmented Generation (RAG) customization where applicable

• Ship measurable improvements to developer productivity (e.g., PR cycle time, CI duration, AI-assisted PR throughput) • Write clean, well-tested code with thoughtful unit and integration coverage • Contribute to Hyperproof's internal developer documentation, CLAUDE.md skill files, and MCP server configurations • Work with Claude Code and GitHub Copilot day-to-day, and help evaluate tools like CodeRabbit for automated code review and PR summarization • Help build automated agents for customer bug triage, PR regression classification, and error-to-fix workflows that chain multiple AI agents together • Author and maintain CLAUDE.md skill files, MCP server configurations, and prompt engineering improvements • Contribute to test impact analysis, flake quarantine, pipeline parallelization, and build acceleration • Help provision multi-environment dev setups, authenticated MCP servers, and connected documentation that enable agentic coding workflows • Build dashboards and instrumentation tracking PR throughput per developer, Claude usage rates, code coverage, and developer experience sentiment • Explore using AI to automatically diagnose production errors from telemetry and propose or generate fixes • Alongside AI-focused projects, get scoped feature and bug-fix work in TypeScript/Node.js/React and Java/C#

• Design, build, and maintain real-time AI-powered APIs and services in Python. • Architect, optimize, and deploy LLM workflows leveraging LangGraph or similar orchestration frameworks. • Develop agentic logic and reasoning pipelines for conversational and task-based automation. • Build scalable infrastructure to support multi-model, low-latency, and streaming AI experiences. • Collaborate with backend and product teams to translate complex customer workflows into reliable AI systems. • Ensure high performance, reliability, and data security across production workloads. • Deploy and manage applications on modern cloud environments (AWS / GCP). • Implement monitoring, logging, and fault-tolerance mechanisms for production-grade AI systems. • Contribute to architectural and product decisions, balancing innovation, safety, and scalability.

• Design, build, and ship production-grade AI agents for meditation, wellbeing, mental health, sleep, anxiety, stress, and behavior change. • Build agentic workflows that combine technical reliability with clinical and evidence-informed reasoning. • Integrate AI agents with backend APIs, data systems, content systems, and product surfaces. • Build evaluation, testing, and monitoring systems for reliability, safety, quality, and continual improvement. • Design clinical and wellbeing evaluation rubrics for AI outputs, including clarity, usefulness, safety, evidence alignment, and risk. • Work with the Head of AI, clinical research team, product, and engineers to translate clinical and research requirements into product and system design. • Identify risks in AI-generated wellbeing outputs, including unsupported claims, unsafe recommendations, overstatement, inappropriate advice, and poor handling of vulnerable users. • Improve prompts, tools, agent workflows, and retrieval systems based on both engineering performance and clinical/product quality. • Architect flexible, modular workflows that make it easy to test, extend, and improve AI capabilities. • Help define when agentic systems should answer, ask for more context, escalate, refuse, or redirect. • Design data and feedback pipelines that improve agent performance through real-world usage, evaluation, and, where appropriate, fine-tuning. • Translate early AI prototypes into scalable, maintainable product systems used by millions of people. • Help shape Insight Timer’s internal standards for building safe, evidence-informed AI wellbeing products.