MediaMint
Website:
mediamint.com
Job details:
Job Description: Head of Agentic Operations owns the complete operating system: intake → prioritization → build → deploy → monitor → improve. You establish the factory that ships governed, observable, repeatable agentic workflows at enterprise scale — enabling engineers to focus on platform evolution while SMEs focus on client value. This is not a back-office operations role. You are the face of our agentic capability to enterprise clients, the architect of delivery excellence across 150+ accounts, and the technical leader who ensures every agent we ship meets production-grade standards.
What You’ll Own:
1. Client-Facing Partnership & Advisory Serve as the senior operational voice in client engagements, translating business pain into agentic solutions: ● Client discovery & solutioning: Lead intake sessions with enterprise clients (Netflix, Disney, Amazon, Cox, NYT, Dow Jones) to understand workflows, define agentic workflow opportunities, and scope agent requirements ● Stakeholder management: Present agent roadmaps, performance dashboards, and ROI narratives to VP/C-level client stakeholders — owning the "what we're building and why" conversation ● Expectation engineering: Set and manage SLAs, success criteria, and acceptance thresholds collaboratively with clients — not just internally ● Escalation ownership: Serve as the senior escalation point for agent-related production issues impacting client workflows; lead war rooms and communicate resolution with clarity and accountability ● Trusted advisor posture: Build durable relationships that position MediaMint as a strategic partner, not just a vendor — proactively surfacing agentic workflow opportunities clients haven't identified
2. Delivery Excellence at Scale
Own end-to-end delivery quality across the agent portfolio — from first commit to production steady-state: ● Delivery cadence: Establish and enforce 3–7 day sprint cycles with clearly defined gates: spec review → build → test → stage → deploy → validate ● Multi-client orchestration: Manage delivery across 150+ client accounts simultaneously - balancing competing priorities, resource allocation, and timeline commitments without dropping quality Golden path enforcement: Define the "right way" to ship an agent - standardized templates, acceptance criteria, golden datasets, success metrics — and hold teams accountable to it Outcome tracking: Own delivery KPIs: on-time deployment rate, first-pass success rate, time-to production, defect escape rate, agent adoption metrics ● Continuous improvement: Run blameless retrospectives, track action items to closure, and feed learnings back into templates and playbooks — the system gets better every sprint
3. Technical Leadership & Architecture Governance
Provide hands-on technical direction without being the one writing code: ● Modular architecture governance: Enforce Lego-style composability across the agent library - prevent one-off sprawl, ensure reusable modules (chat Q&A, knowledge graphs, media plan generators) are the default ● Release governance & evaluation pipelines: Own agent release criteria, quality gate definitions, and evaluation pipelines — golden dataset curation, behavioral validation thresholds, cost guardrails, and client-specific rollout orchestration — leveraging platform-embedded CI/CD infrastructure Quality engineering for non-deterministic systems: Build trust in GenAI-powered workflows through golden test sets, behavioral validation, statistical thresholds, accuracy drift detection, and hallucination guardrails ● Performance standards: Define and enforce latency budgets, throughput SLAs, inference cost targets per agent type, and prompt injection defenses ● Technology radar: Evaluate and recommend tooling decisions — LLMOps platforms, evaluation frameworks, browser automation, observability stacks — and drive build-vs-buy recommendations with data ● Platform investment prioritization: Work with Engineering to sequence infrastructure investments (tooling, APIs, SDKs) based on delivery impact and client urgency
4. Production Reliability & Observability
Establish enterprise-grade operational discipline: ● Real-time monitoring: Agent success rates, failure modes, drift detection, cost-per-run analytics ● Alerting & escalation: PagerDuty integration, on-call rotations, severity classification ● Incident management: War rooms, blameless postmortems, preventative action tracking ● Capacity planning: Forecast compute/LLM costs vs. agentic workflow ROI; optimize resource allocation
5. Cross-Functional Execution Leadership
Translate operational pain into shipped outcomes: ● Partner with Product to align agent roadmap to client needs and business KPIs ● Work with Engineering to prioritize platform investments (tooling, infrastructure, APIs) Collaborate with SME teams to capture tribal knowledge and codify it into reusable agents Interface with Delivery to ensure smooth handoffs and production adoption ● Coordinate with Sales & New Business to support enterprise deal cycles with technical credibility, pilot scoping, and reference architecture narratives
6. Standardization With Flexibility
Enable scale without becoming a bottleneck: ● Core platform standards: Authentication, logging, error handling, cost tracking ● Client variation framework: Parameterized configurations vs. hard-coded customizations Self-service tooling: Empower SMEs to extend agents within governed boundaries ● Documentation culture: Runbooks, decision logs, architectural decision records (ADRs)
. Team Building & Leadership
Develop a high-performing, learning-oriented team: ● Hire and onboard: Attract talent with agentic workflow platform, LLMOps, or enterprise ● DevOps backgrounds Mentorship: Grow Technical PMs into agentic systems architects; elevate QA into AI evaluation specialists ● Performance management: Define career ladders, run 360 reviews, drive continuous skill development ● Culture: Establish operational discipline, blameless learning, ruthless prioritization
Required Qualifications:
Must-Have Experience ● Leadership at Scale: 10+ years in engineering, automation, or delivery leadership; proven track record building and managing teams (10+ direct/indirect reports) in multi-client environments ● Client-Facing Delivery: Demonstrated experience presenting to VP/C-level stakeholders, managing SLAs, navigating escalations, and building trusted advisor relationships with enterprise clients ● Production Operations: Owned enterprise-scale systems serving business-critical workflows (SLA driven environments, multi-client SaaS or services operations) ● Automation / Agentic Workflow Platforms: Built or scaled orchestration systems, workflow automation COEs, agentic workflow factories, or integration-heavy platforms ● Operational Rigor: Defined release governance criteria, quality gates, and evaluation frameworks; experience with CI/CD pipelines, monitoring/alerting, incident response, and measurable reliability improvements ● Metrics-Driven Execution: Defined KPIs, managed teams to outcomes (not activity), demonstrated a continuous improvement culture
Preferred Experience:
● Experience in analytics consulting, digital operations, or ad-tech/media-tech services environments. ● Familiarity with GenAI/LLM production systems, agentic frameworks, or AI evaluation methodologies. ● Background in managing delivery across multiple concurrent client accounts (5+) ● Exposure to SOC 2 compliance, enterprise security questionnaires, or regulated delivery environments
Click on Apply to know more.