• Collaborating with Data Science and Product teams to design, build, and ship LLM-powered features across our media buying platform. • Designing and building agents using frameworks including LangChain and LangGraph to automate complex media planning workflows. • Developing MCP servers and APIs that connect internal and external agents to our media buying toolkit. • Building and tuning agent orchestration workflows, including evaluation and observability, to ensure agents reason, plan, and execute reliably. • Rapidly experimenting with emerging agentic approaches while guiding stable, scalable architectural decisions.

• Help train large-language models (LLMs) to write production-grade code • Compare & rank multiple code snippets • Repair & refactor AI-generated code • Inject feedback into the RLHF pipeline

• Work on services supporting our customer-facing web and mobile apps • Work mostly on the backend but with a full-stack mindset. • Use AI tooling for most, if not all, of your tasks • Write your unit tests • Automate stuff. Like a lot. • Be religious with Git • Depend on a CI pipeline and world-class telemetry. • Experiment, learn, and have fun

• SRE Leadership & Strategy: Set clear goals for the SRE team and partner with Engineering leadership to align platform initiatives with business objectives. • Reliability & Observability (SLA/SLO): Lead the definition and enforcement of SLAs, SLIs, and SLOs. Architect observability frameworks to translate telemetry data into actionable roadmaps that reduce toil and enhance resilience. • Core Engineering & Performance: Take ownership of critical code components (i.e., Queues, Enrollments) and lead efforts to identify bottlenecks, optimize performance, and improve code quality across the engineering department. • Security by Design: Champion infrastructure security. Partner with InfoSec to define hardening standards, manage perimeter defense (WAF/DDoS), and automate vulnerability remediation within the CI/CD pipeline. • Incident Command: Participate in the 24x7 on-call rotation and lead post-incident reviews (RCAs), ensuring action items are implemented to improve MTTR and prevent recurrence. • Mentorship: Empower developers with better tooling and guidance on performant coding practices, fostering a culture of collaboration and reliability and "you build it, you run it".