About Hupo

We are an AI-native start-up building sales enablement products in the banking, financial services, and insurance industry; already trusted by dozens of enterprise customers (including Fortune 500 companies).

Role at a Glance

Title: Product Manager
Team: Product
Employment Type: Full-time
Location: Remote (UTC+5–UTC+11 / Asia time zone preferred)

Why This Role Matters

Hupo is building AI products that help enterprise banks and insurers improve how their frontline teams identify opportunities, make decisions, and serve customers.

This role will take Hupo's agentic automation capabilities from concept to a product line. You'll work at the intersection of AI, enterprise workflows, and regulation — building products that can make increasingly autonomous decisions while ensuring they remain trustworthy, governed, and appropriate for highly regulated institutions.

The decisions you make in this role will help define how Hupo builds and deploys agentic AI across its product suite.

What You'll Do

Take agentic automations from concept to a product line. Own roadmap, prioritisation, OKRs, and outcomes for decisioning capabilities: lead categorization and scoring, lead-to-advisor matching, renewal and cross-engagement opportunity detection.
This is 0 to 1 work with live enterprise clients, and the calls you make now will define these products.
Own the governance harness for agentic AI. Define how every agentic capability we ship is evaluated, evidenced, and certified before it acts on behalf of a regulated institution: autonomy boundaries, human-in-the-loop requirements, audit trails, and evaluation gates. You set the bar; product owners across the suite build to it. You will work directly with our compliance and engineering leadership on this.
Design autonomy within regulation. Our clients are banks and insurers across Singapore, Malaysia, Taiwan, and Hong Kong.
Deciding what a system may do autonomously, what requires a licensed human in the loop, and how that decision is evidenced is core product work here, not an afterthought.
Build on shared infrastructure. Our agentic products run on a common runtime alongside our conversational agents. You will define the capabilities your automations need and work closely with the peer PM and platform engineering, rather than building a parallel stack.
Go deep with enterprise BFSI customers. Understand distribution workflows, agency structures, and compliance constraints firsthand, and translate them into products advisors and their institutions rely on daily.
Lead a squad. Ship and iterate fast with a team of engineers and a designer, without cutting corners on evaluation or safety.
Operate AI-native. Use AI tooling by default to multiply your own output, and show us how you do it.
Use data to define success, track progress, and keep raising the bar, including for the automations themselves: match quality, precision of opportunity detection, advisor adoption, and downstream conversion, not just usage.

What Success Looks Like

Within your first few months, you'll:

Take ownership of key agentic automation capabilities and establish clear priorities, outcomes, and measures of success.
Develop a deep understanding of Hupo's enterprise BFSI customers and translate their workflows and constraints into product decisions.
Establish strong foundations for evaluating and governing agentic capabilities as they move from concept into real-world deployment.
Ship and iterate on products alongside Engineering and Design while maintaining a high bar for safety and evaluation.

You’ll also be expected to continuously:

Make thoughtful decisions about where AI should operate autonomously and where human involvement is required.
Raise the standard for how Hupo evaluates and governs agentic AI.
Use data and customer outcomes to continuously improve the quality and impact of Hupo's automation capabilities.

What We're Looking For

Must-Haves

We hire product builders who are already ahead of their peers. For this role, the constants:

You have built LLM-powered decisioning or automation systems in practice: tool use, orchestration, structured outputs, and evaluation design, and you have shipped something real with them.
You know how to make an AI system's decisions trustworthy: evaluation harnesses, guardrails, human review workflows, and audit evidence are things you have built, not concepts you have read about.
You have sound judgment about autonomy in high-stakes settings. You can articulate when a system should act, when it should recommend, and when it must hand off to a human, and you can defend that reasoning to a compliance officer.
You are fluent in the underlying stack: prompting, retrieval, and evaluation are working tools for you, not vocabulary.
You reach for AI tooling by default to compress your own work, and you can demonstrate it.
You thrive in ambiguity and take ownership without waiting to be told what to do. There is no established playbook for this suite. You will write it.
You communicate clearly and concisely, in writing and in person, and you translate between technical, commercial, and compliance audiences.

Nice-to-Haves

Deep domain knowledge in insurance, wealth management, or financial services sales and servicing.
Enterprise SaaS experience, especially deploying into large regulated institutions.
Experience with lead management, CRM intelligence, sales operations tooling, or recommendation and matching systems.
Familiarity with AI governance frameworks in financial services (for example MAS FEAT, HKMA GenAI guidance, or equivalent).