• End-to-end ML pipelines: from feature engineering to deployment • RAG systems with semantic search and knowledge graphs • Scalable generative AI services in production environments • Robust data pipelines for high-volume workloads

• Design agent behaviours, reasoning flows, interaction patterns, delegation rules, and tool-use strategies • Develop orchestration logic for multi-step, multi-agent, and tool-augmented workflows • Create and refine prompts, system instructions, behavioural frameworks, validation rules, and safety guidelines • Build lightweight prototypes and POCs to validate agent logic and orchestration patterns • Run structured evaluations (reasoning consistency, tool-use correctness, stability) • Identify hallucinations, edge cases, and failure patterns; propose improvements • Collaborate with engineers to validate feasibility and maintainability • Work with architects, engineers, and business teams

• Define architectures for AI solutions (agents, orchestration, RAG/CAG, memory). • Produce architectural diagrams and define non-functional requirements. • Select appropriate design patterns. • Validate designs via reference artifacts or spikes when needed. • Design integrations between AI services and enterprise systems. • Define APIs, contracts, error handling, and authorization models. • Ensure interoperability across systems and platforms. • Provide reusable architecture patterns. • Define retrieval strategies (chunking, embeddings, vector stores). • Promote reusable components and templates. • Define acceptance criteria (accuracy, latency, cost, safety). • Establish evaluation scenarios and quality gates. • Ensure feedback and telemetry loops are included. • Embed security-by-design principles. • Address risks like prompt injection, data leakage, and misuse. • Ensure observability, traceability, and rollback strategies. • Define mitigation strategies for AI threats. • Align solutions with enterprise platforms (Azure OpenAI, etc.). • Ensure compliance with constraints and standards. • Identify gaps and propose solutions. • Make build vs. buy decisions. • Advise stakeholders on feasibility and trade-offs. • Lead design reviews and maintain architectural integrity. • Communicate decisions clearly. • Maintain architecture documentation and ADRs. • Define best practices and reusable standards.

Role Description The AI Product Manager will own our AI product strategy and help us identify and execute the biggest AI opportunities across our products and operations. This is a high-impact, cross-functional role sitting at the intersection of product, data, and engineering. This position will work closely with engineering, data, design, and operations teams to define what we build, validate that it creates real value, and ship features that leverage AI in meaningful, responsible ways. This role will also serve as an internal advocate for AI — helping stakeholders across the company understand where AI can move the needle. Key Responsibilities - AI Strategy & Opportunity Identification: - Conduct structured discovery across our product lines and internal operations to identify the highest-leverage AI opportunities. - Build and maintain an AI roadmap, prioritizing initiatives by impact, feasibility, and alignment with company strategy. - Evaluate emerging AI capabilities (LLMs, generative AI tools, AI APIs) and assess their relevance to our products and users. - Partner with leadership to define a coherent AI vision and communicate it clearly to the organization. - AI Product Execution: - Define clear success metrics for AI products, including accuracy, latency, user adoption, and downstream business impact. - Own the full product lifecycle for AI features — from discovery and scoping through launch and iteration. - Work hands-on with engineering teams to manage trade-offs, write specs, and unblock delivery. - Build safeguards and feedback loops to catch bad AI outputs, edge cases, and misuse in production. - Lead post-launch reviews to drive continuous improvement of AI features. - Data Analysis & Insights: - Analyze product and behavioral data to identify opportunities, validate hypotheses, and measure outcomes. - Translate ambiguous signals (flat DAUs, rising support tickets, usage drop-offs) into clear diagnostic frameworks and action plans. - Work comfortably with SQL, BI tools, and data analysts to pull and interpret data independently when needed. - Design experiments and A/B tests to validate AI feature changes rigorously. - Data Platform Partnership: - Collaborate with data engineering teams to understand data availability, pipeline readiness, and infra constraints. - Incorporate data platform considerations into roadmap planning and feature design. - Serve as the product voice in discussions about data architecture and the data signals that power AI features. - Cross-functional Leadership: - Build alignment across engineering, design, operations, legal, and leadership on AI priorities and timelines. - Proactively identify and resolve organizational friction that slows down AI initiatives. - Communicate clearly about AI product decisions — what we're building, why, and what tradeoffs we're making. - Perform additional tasks as needed to support the success of the company. Qualifications - 3+ years of experience in product management or a related field. - 2+ years shipping AI-powered products. - Demonstrated ability to identify and validate new product opportunities, especially in AI domains. - Strong data analysis skills — comfortable writing SQL, working with BI tools (Looker, Tableau, Metabase, etc.), and running analyses independently. - Experience partnering with data engineering and data science teams to ship data-intensive products. - Excellent written and verbal communication and able to explain complex AI concepts to non-technical stakeholders. - Track record of driving cross-functional alignment on ambiguous, high-stakes initiatives. Preferred Qualifications - BS or BA in Computer Science or related field or equivalent experience. - Experience building LLM-powered products or GenAI features using third-party AI APIs (e.g., prompt engineering, RAG, output evaluation). - Working knowledge of data platforms and tooling (e.g., Snowflake, dbt, Databricks, Airflow, or equivalents). - Familiarity with AI quality frameworks — how to evaluate AI output quality, set up evals, monitor for regressions, and manage hallucination risks. - Hands-on experience with Python for data exploration or lightweight analysis. - Experience in a company that operates both a user-facing product and internal operational workflows. - Background in or exposure to responsible AI practices — fairness, explainability, user trust. Special Requirements - Willingness to travel as required. - Prolonged periods sitting at a desk and working on a computer. - Must be proficient in the English language, both written and verbal. - Ability to lift 15 pounds. Benefits