Lead Technical Product Manager – Generative AI

Lead the integrated plan for work that spans multiple modules; align product, engineering, and UX to support rapid GTM

Transform epics into clear, INVEST features and user stories (Independent, Negotiable, Valuable, Estimable, Small, Testable) with precise acceptance criteria and Definition of Ready/Done

Ensure voice of customer and market data flows into sprint planning and backlog prioritization; translate customer feedback into actionable user stories

Maintain a prioritized backlog in Azure DevOps Boards with 2-3 sprints of refined, ready work, visible dependencies, and unblocked paths to delivery

Apply lightweight prioritization methods (value, risk, effort, sequencing, cost of delay) with documented rationale

Lead backlog refinement sessions, sprint planning, and story elaboration with development teams

Partner with Engineering on slicing, technical feasibility, release planning, feature flags, and canary rollouts

Collaborate with Scrum Master to optimize team flow metrics, maintain predictable delivery, and remove impediments

Apply eXtreme Programming (XP) practices where appropriate, including test-driven development support

Specify product requirements for Azure OpenAI-based features, including grounding to authoritative sources, citation behavior, refusal/abstain rules, and graceful error handling

Understand customer workflows and jobs-to-be-done to effectively decompose AI-driven solutions into implementable features; identify where automation/AI can deliver value within existing user journeys

Collaborate on RAG requirements: content sources, chunking strategy, embedding selection, vector search, retrieval approach, and evaluation criteria

Define AI-specific acceptance criteria and SLOs: groundedness/relevancy, quality thresholds, latency budgets (sub-3s), concurrency, and cost per interaction

Coordinate prompt templates, model change control, and safety guardrails so demos, pilots, and production remain predictable

Work with engineering to define fallback strategies and error handling for AI features

Establish evaluation metrics including performance benchmarks (latency, accuracy, groundedness)

Run short build-measure-learn loops with focus on validated outcomes, not output volume

Design and execute rapid validation experiments to test hypotheses about user needs and solution viability

Define problem-solution fit and product-market fit that maximize learning with minimal development effort

Convert discovery signals and pilot feedback into backlog updates quickly; retire low-value items and reduce WIP

Track innovation metrics including time-to-validation, experiment velocity, and learning rate

Support A/B testing and feature flagging strategies for controlled rollouts

Apply lean startup principles to reduce waste and accelerated validated learning

Coordinate with Product team for customer sessions; capture technical requirements and implementation considerations from these discussions

Coordinate with GTM lead to ensure engineering deliverables align with launch requirements; facilitate knowledge transfer to Sales, Support, and other internal teams pre-release

Support Product Managers in discovery by turning problem insights into hypotheses and testable stories

Integrate user feedback, analytics, and support signals into prioritization; ensure each story anchors to real user problems

Partner with UX on flows that feel intuitive and require minimal training

Work horizontally with platform, security, compliance, and content teams to meet privacy, safety, and auditability expectations

Produce concise artifacts that reduce ambiguity: story maps, acceptance test outlines, release notes, known limitations

Keep stakeholders aligned with short, factual updates: current focus, what shipped, what we learned, what's next

Partner with Scrum Master to maintain dashboards for delivery and product health: throughput, cycle time, story readiness, escaped defects, AI quality and latency

Tie backlog items to measurable outcomes and close the loop with post-release verification

Track and report on key AI metrics including model performance, user adoption, and business impact

5-7+ years in software product management or product ownership in B2B SaaS environments

4+ years practicing Agile/Scrum in Product Owner or Lead PM capacity, working closely with engineering

2+ years working with AI/ML products, with hands-on experience shipping Generative AI features in production strongly preferred

Experience with lean product development and build-measure-learn methodologies

Demonstrated experience in startup environments or innovation labs preferred

Expert backlog hygiene in Azure DevOps Boards: epics to features to stories, acceptance criteria, Definition of Ready/Done, dependency tracking, release planning

Deep understanding of generative AI concepts including LLMs, RAG architectures, prompt engineering, embeddings, and vector databases

Working knowledge of Azure OpenAI Service, prompt patterns, evaluation approaches, and safe response behavior

Strong grasp of INVEST principles and story mapping techniques

Understanding of API integrations and microservices architectures

Knowledge of AI evaluation metrics, testing strategies, and MLOps practices

Understanding of data privacy, security, responsible AI, and auditability in enterprise environments

Problem-first, customer-obsessed, and evidence-driven mindset

Self-starter mentality with ability to work independently in ambiguous environments

Critical thinking skills to challenge assumptions, simplify complex requirements, and validate hypotheses

Exceptional written and verbal communication for technical and non-technical audiences

Comfort with rapid iteration and ability to pivot based on learning

Strong facilitation and conflict resolution skills

Clear, direct communicator who collaborates well across functions

Certified Scrum Product Owner (CSPO/PSPO) or SAFe POPM certification

Azure AI-900 or AI-102 certification

Background in professional services software (tax, accounting, legal)

Experience managing distributed or remote development teams

Familiarity with document intelligence technologies

A transparent, prioritized backlog with 2-3 sprints of ready stories and minimal rework

Shipped GenAI capabilities that meet acceptance criteria for grounding, safety, latency, and usability

Faster learning cycles, fewer blocked items, and clear evidence that shipped work solves real user problems

Short, useful updates that keep stakeholders aligned without ceremony overhead

Consistent delivery with decreasing cycle times and increasing customer value