VP of AI, ML and Data Strategy

Shipium

6h•$277,000 - $290,000•Remote

About The Position

This leader heads Shipium's Data Science, AI, Data Engineering, and Business Intelligence Engineering functions — and the production systems they own run the platform. The team owns the production machine learning behind delivery promise, carrier selection, fulfillment optimization, and simulation, including real-time inference at platform scale. It owns the data engineering platform that powers analytics and modeling — the warehouse, transformation pipelines, and orchestration that downstream systems depend on. And it owns Business Intelligence Engineering, which delivers Orca Analytics and the internal and customer-facing reporting Shipium runs on. The function is also a foundational partner to Shipium's broader Generative AI work, providing the data, LLM, and agent infrastructure, evaluation practices, and proprietary model integrations that other product and engineering teams build on. The VP balances executive-level strategy with substantive technical engagement — an architecture review, a model post-mortem, or hands-on contribution to the codebase when the work demands it. Success rests on technical credibility, an exacting hiring bar, and sound judgment across the organization. The role works across Product, Engineering, Implementations, Sales, Marketing, and Finance to turn these capabilities into customer outcomes.

Requirements

An AI-native practice: This leader is expected to use Claude and Claude Code as part of daily working practice — not just for drafting documents — and to set the bar for how the team uses AI tooling in its own work. Candidates should view AI as core working infrastructure.
Build reusable agents, skills, or prompt patterns for personal and team leverage.
Automate multi-step research, documentation, and analysis pipelines.
Work with MCP-based integrations to connect AI tools to internal systems (Datalake, Jira, Confluence).
Use AI to scale individual leverage rather than replace technical judgment.
Evaluate AI tools across the function with the same rigor applied to any technical investment — for individual productivity, engineering velocity, and the team's own output.
10+ years in data science, machine learning, or data engineering, with at least 5 years in a people-management or technical leadership role — including managing managers or tech leads and representing the function to executive stakeholders.
Demonstrated experience building or substantially scaling a data, ML, or AI organization — hiring senior individual contributors and leaders, standing up or maturing the operating model, and owning a function-level budget.
Strong working knowledge of the modern ML and data stack, with credible architectural and investment judgment across all of it. ML frameworks (e.g., scikit-learn, PyTorch, XGBoost) and applied experience with production predictive modeling.
LLM application development — orchestration frameworks (e.g., LangChain, LlamaIndex), prompt engineering, evaluation harnesses, RAG, and tool-using agents; familiarity with commercial and open models (e.g., OpenAI, Claude, Gemini, Llama) and their trade-offs.
Modern data platforms — data warehousing on Amazon Redshift, transformation via dbt, orchestration via Airflow, and data lakes on AWS S3.
Cloud infrastructure on AWS — including SageMaker, S3, and Bedrock — and a working sense of the cost and reliability trade-offs across managed vs. self-hosted options.
Production ML operations — CI/CD for ML, model monitoring, and observability (e.g., Datadog).

Nice To Haves

B2B SaaS, supply chain, logistics, or e-commerce experience strongly preferred.
Master's degree in Computer Science, Data Science, Statistics, Mathematics, Operations Research, or a related quantitative field preferred. Equivalent industry experience considered.

Responsibilities

Set the multi-year strategy for ML, data, and BI, grounded in customer signal, competitive intelligence, platform reality, and financial constraint — and position the function's Generative AI work as a foundational capability other teams build on.
Connect ML, data, and BI investments into a coherent narrative about how Shipium turns operational data into customer outcomes, and communicate that narrative clearly to the executive team.
Make build-vs-buy and managed-vs-self-hosted calls across the ML and data stack, with a clear rationale documented for the team and for finance.
Run quarterly and annual planning for the function, operating the roadmap as a structured artifact reconcilable with Jira and stakeholder communications.
Build and scale the Data Science / AI, Data Engineering, and BI Engineering organization — owning org design, headcount strategy, and the leveling that takes the function to its next stage of maturity.
Manage managers and tech leads, not only individual contributors; grow the internal leadership bench and create the management layer the org needs as it scales.
Own the technical hiring bar across all three sub-functions. Run senior and leadership recruiting personally; design the interview loops and calibration that protect the bar as hiring volume grows.
Set and enforce engineering and analytical standards — code quality, testing, monitoring, documentation, and model-evaluation discipline — through review and operating cadence rather than exhortation.
Run performance management, career development, and retention; define leveling and progression for data science, ML, data engineering, and BI engineering tracks.
Foster a culture of technical rigor, written communication, and cross-team ownership, and act as its visible standard-bearer.
Operate as a peer to VP-level leaders across Product, Engineering, Implementations, Sales, Marketing, and Finance — owning the data/ML/AI position in executive decisions rather than informing them from the side.
Partner with Product on roadmap and prioritization so modeling and analytics ship as part of the product; jointly own the outcomes, not just the models.
Partner with Engineering leadership on deployment, platform integration, on-call, and incident response for production models and pipelines.
Partner with Implementations and Sales on the data and AI capabilities surfaced pre-sale, during onboarding, and across the customer lifecycle; serve as VP-level technical presence in strategic customer conversations.
Partner with Finance on budget and the unit economics of LLM and cloud usage; own the function's spend and the build-vs-buy economics behind it.
Serve as Shipium's senior technical voice on data and AI in market-facing work — case studies, thought leadership, and analyst and customer conversations — and represent the function as a standing participant in company strategy.
Own the architecture, deployment, and operation of the production ML behind delivery promise, carrier selection, fulfillment optimization, and simulation, including the real-time inference systems that serve them at platform scale.
Own the data platform underneath analytics, modeling, and product features: data lake on AWS S3, warehouse on Amazon Redshift, orchestration via Airflow, transformation via dbt, and the training-data and feature pipelines downstream ML depends on.
Own Business Intelligence delivery, including Orca Analytics and the customer-facing analytics surfaces the BI team builds.
Drive disciplined model-lifecycle practice — retraining cadence, evaluation thresholds, shadow/production validation, and registry and release-notes hygiene.
Establish monitoring, alerting, and incident-response practice appropriate to a production platform (e.g., Datadog), and own the operational quality bar and SLAs for the models and pipelines that other Shipium products depend on.
Set the reliability and cost trade-off posture across managed vs. self-hosted infrastructure.
Lead the function's Generative AI work as a foundational capability — the data, LLM, and agent infrastructure, orchestration, and evaluation practice that other teams build on, rather than a standalone product surface.
Set the technical direction for how Shipium's proprietary predictive models integrate with LLMs to power agent and assistant capabilities.
Own build-vs-buy and managed-vs-self-hosted calls for LLM providers, vector stores, orchestration frameworks, and agent infrastructure — mapping each use case to strategic differentiation, data sensitivity, and 3-year TCO, and revisiting those commitments on a deliberate schedule.
Establish evaluation, benchmarking, validation, and responsible-AI practices — model evaluation, output validation, and regression harnesses — and push back on unbenchmarked or generic GenAI patterns.
Own the external partnerships relevant to Generative AI (cloud and model providers), including the commercial and roadmap relationships.