AI Product Manager

Mulligan Funding; San Francisco (CA); On-site; USD 120,000 - 190,000; Full time

Job summary

Mulligan Funding is hiring an AI Product Manager to embed in core AI initiatives. You will lead discovery and design, sit with SMEs to map workflows, and partner with engineering to produce production-grade AI solutions.

The role demands comfort in ambiguity and strong collaboration with business teams to drive fast, AI-driven decisioning. You will report to the AI Product Director and work within a growing, agile pod, expanding AI capabilities across Sales, Customer Lifecycle, Finance, and

Qualifications

• Experience leading discovery and design for AI initiatives.
• Ability to map current workflows with SMEs and redesign them using AI tools.
• Comfort operating in ambiguous environments and turning messy problems into clear designs.

Responsibilities

• Lead discovery sessions with SMEs to map current processes, pain points, and cycle times.
• Translate findings into an AI transformation design brief and acceptance criteria.
• Manage day-to-day collaboration with engineering partners and ensure alignment to the design brief.
• Design and implement a QC framework, evaluation rubrics, and production-ready outputs.

Skills

AI product management

Discovery sessions

Stakeholder management

Partner management

Process mapping

Go/No-Go decisions

Job description

Headquartered in San Diego, Mulligan Funding serves as a leading provider of working capital (Up to $5M) to the small and medium-sized businesses that fuel our country. Since 2008, we have prided ourselves on our collaborative, innovative, and customer-focused approach. Enjoying a period of unprecedented growth, driven by the combination of cutting-edge technology, human touch, and unwavering integrity, we are looking to add to our people first culture, with highly motivated and results-oriented professionals, to push the limits of what’s possible while creating value for all of our partners.

At Mulligan, we are replacing legacy small business lending with fast, intelligent, AI-driven decisioning. Backed by 18 years of proprietary credit data and deep risk expertise, we have the institutional knowledge and technical foundation that competitors simply can’t match. We are looking for builders who want to be right at the center of this work.

We’ve already moved past the experimental phase. Production-grade AI agents are actively running in our credit and underwriting workflows today, driving clear results in speed, consistency, and decision quality. Now, we are expanding that AI-first approach across Sales, Customer Lifecycle, Finance, and Capital Markets.

We are hiring an AI Product Manager to embed directly inside one of our core AI initiatives. In this hands-on role, you will lead discovery and design for active AI projects, work shoulder-to-shoulder with business teams, and manage execution alongside specialist engineering partners.

We need someone who thrives in ambiguous environments and can take a messy problem and turn it into a clear design. Your main focus on day one will be sitting with functional experts, mapping out how their daily workflows operate, and redesigning them from scratch using modern AI tools.

Reporting to the AI Product Director, you will operate with high day-to-day autonomy inside your pod. This position is ideal for an early-career product manager looking to build deep AI expertise fast and grow alongside our expanding team.

Current State Discovery & Process Intelligence

• Deep-Dive SME Discovery: Lead structured discovery sessions with functional Subject Matter Experts (SMEs) to map the current process in full — every step, decision point, pain point, exception case, volume estimate, and cycle time. The goal is to understand how the process actually works, not how it is documented.
• Metric Baselining: Quantify the current state — cycle times, error rates, manual touchpoints, cost of failure — to create the measurement foundation against which post-deployment impact will be assessed.
• Edge Case Extraction: Surface the process edge cases, informal workarounds, and undocumented exceptions that SMEs carry in their heads but rarely write down. These are often where AI creates the most value — and where designs fail if not addressed.
• Hypothesis Testing: Work with the AI Product Director to test design hypotheses during discovery — some will be confirmed by the process reality, some will need to be revised. Discovery should challenge assumptions, not confirm them.

AI Solution Design & Brief Authorship

• 0-to-1 Reimagining: Translate current state intelligence into an AI transformation design brief that reimagines the process from scratch — not an AI-assisted version of the status quo. The question to answer is: if we were designing this process from scratch with AI available, what would it look like? Vision and ability to think outside of the box is mandatory
• Design Brief Production: In close collaboration with the AI Product Director, author a detailed solution design brief that defines the AI capability, its inputs and outputs, the logic it applies, the QC framework it requires, and the acceptance criteria the engineering partner must meet. This brief is the specification the partner builds to — precision matters.
• Partner Identification: Conduct research to identify specialist engineering partners with a demonstrated track record in the specific use case domain. Apply Mulligan’s IP sensitivity framework to determine the appropriate engagement model.
• Go/No-Go Recommendation: Produce a clear go/no-go recommendation to the AI Product Director with supporting rationale — including current state baseline, proposed AI solution design, recommended partner, and initial cost-benefit framing.

Engineering Partner Management

• Day-to-Day Partner Interface: Manage the working relationship with the specialist engineering partner through the build cycle — coordinating build reviews, tracking progress against the design brief, and keeping delivery on schedule.
• Deliverable Review: Review partner outputs against the design brief and Mulligan’s acceptance criteria — identifying divergences early and escalating material gaps to the AI Product Director before they become rework cycles.
• Iterative Build Cycle: Facilitate the review-feedback-refine loop between the engineering partner and Mulligan’s internal QC and functional stakeholders — ensuring the build converges on a production-grade output, not a minimum viable approximation.
• Vendor Discipline: Enforce Mulligan’s core partnership principle: Mulligan authors the specification; the partner builds to it. Never allow the partner to define the scope or shape the design.

QC Framework & Evaluation Support

• Pre-Build QC Design: Design the QC framework for the pod’s use case during Phase 1 — before the engineering build begins. Define the human review process, the deterministic checks the DS person will build, and the LLM-as-judge rubric that will evaluate output quality.
• Baseline Evaluation: Collaborate with the pod’s DS person to establish evaluation rubrics, golden datasets, and accuracy thresholds that define what ‘production ready’ means for this specific use case.
• First Impression Discipline: Ensure the first outputs from any AI agent are