Staff Quality Engineer - Exploratory Test Lead

Bestow is a leading vertical technology platform serving some of the largest and most innovative life insurers. Our platform unifies the fragmented, legacy value chain, enabling carriers to launch products in weeks instead of years. We build the infrastructure that helps the life insurance industry move faster, reach more people, and deliver on its promise. Backed by leading investors and trusted by major carriers, Bestow is powered by a team that moves with precision, purpose, and heart.

About The Team

Bestow's Quality Engineering organization is building the verification backbone for an AI-enabled software development lifecycle: specification-driven development, Definition-of-Ready enforcement, requirements-to-test traceability, and a growing non-functional verification stack. We shape how software is defined, not only how it is checked.

About The Role

This Staff Quality Engineer role is embedded first within Bestow Innovation Lab, the R&D group that invents the next generation of our platform and then scales the practice across Engineering. You will work alongside domain experts and researchers, product engineering teams, Platform Engineers, Technical Program Management, Compliance, and the rest of the Quality Engineering team. This role reports to the Director of Quality and is open to Remote (US).

How We Think About Testing

• Test automation and automated checks confirm what we already knew to ask, but they cannot tell us whether we asked the right questions.
• Testing is the process of evaluating a product by learning about it through exploration and experimentation; questioning, modeling, observation, inference, and output checking.
• We distinguish testing from checking. A check applies a decision rule to an observation and reports the outcome. Checks are enormously valuable and we invest heavily in them, but they cannot substitute for a person trying to find out what is true.
• Tools extend the tester. We expect you to build and use tools aggressively, including AI technologies, for data generation, state manipulation, differential comparison, log and trace analysis, and oracle construction.
• Testers inform decisions; they do not own quality. Our north star is to give the people accountable for a release the clearest possible account of the product and its risks, including what was not examined.
• Risk analysis is an argument, not arithmetic. We want risk stated as a specific failure story that colleagues can inspect and dispute.
• We are careful about what we claim. "So far I haven't found a problem here" and "this area is fine" are very different statements, especially in a regulated business.

Responsibilities

Testing, Embedded in Labs / R&D (~40%)

• Embed with Labs / R&D through invention and specification, developing a model of the product as it is being conceived, capturing risks and questions before they become code.
• Design and execute chartered, time-boxed testing sessions. Through structured debriefs, capture vital learnings, refine session charters, and validate the return on testing efforts.
• Investigate critical failure modes that structural checks cannot catch: multi-tenant isolation, configuration interaction, concurrency and lifecycle state transitions, integration timing, operability and observability gaps, and long-horizon temporal behavior such as renewals, lapses, and mid-term endorsements.
• Model risk using the Heuristic Test Strategy Model; product elements (structure, function, data, interfaces, platform, operations, time), quality criteria, and risk heuristics. Maintain a living risk catalogue across policy lifecycle, claims, partner integrations, and regulatory workflows.
• Report findings in four distinct kinds:
  • Bugs – anything about the product that threatens its value, filed with severity, priority, and investigative context.
  • Risks – plausible failure stories not yet demonstrated, which become charters.
  • Questions about intent – missing, ambiguous, or contradictory rules, sent back to specification.
  • Obstacles to testing – testability limits, missing observability, or tools you don’t have.
• Maintain a coverage and testability map: what we have examined, with respect to what model of the product, and where we currently cannot see.
• Practice bug advocacy: getting important problems understood and fixed by people who are proud of what they built.

Specification & Risk Engagement (~25%)

• Partner with Product, Labs, and Engineering in Example Mapping and Specification-by-Example workshops to sharpen "Definition of Ready" by surfacing missing rules, unexamined assumptions, conflicting behaviors, and persona variation.
• Bring the oracle question into specification work: how would we know this behavior was wrong? Requirements that cannot be tested against any oracle are not yet requirements.
• Analyze and advocate for system testability through deep observability, precise controllability, and trustworthy test oracles. Escalate lack of examinability as a core defect.
• Identify where automated checks cannot detect a class of failure, and route those to the automation and specification-test backlogs.
• Aim for sufficiency for the decision at hand rather than completeness. Make investigation part of the traceable record for audits.

Tools, Automation & AI Under Test (~15%)

• Build and use tools to extend your reach: data generation, state manipulation, differential and round-trip comparison, log and trace mining, property-based and high-volume techniques, and oracle construction.
• Investigate AI-generated code, tests, and agentic workflow output: prompt-to-implementation mismatch, variation across identical inputs, multi-step state consistency, auditability, and whether generated tests detect regressions.
• Use agentic coding tools as part of your workflow and scrutinize their output as rigorously as any other code.
• Help the organization read its dashboards honestly. A green suite is evidence that specific checks passed, not evidence of quality.

Coaching, Enablement & Decision Support (~20%)

• Build skill in others through coaching, paired sessions, debriefs, and critique of real work. Support with lightweight artifacts: charter catalogues, session notes conventions, heuristics, and oracle references.
• Coach Quality Engineers, SDETs, engineers, and product managers to model risk, charter sessions, take useful notes, and report findings effectively.
• Produce the testing story for release and transition decisions: a story about the product, the testing, and the quality of that testing, including what was not examined.
• Contribute to release readiness practice, promotion gates, rollback validation, and canary strategy as a source of evidence and risk narrative.
• Supply failure hypotheses and scenarios to the non-functional verification stack: performance modeling, resilience, accessibility, security, and chaos experiments.
• Strengthen how we account for testing to regulators and auditors. Escalate compliance findings promptly and precisely.

Requirements

• 10+ years of demonstrated experience with progressively increasing responsibility in software testing and test automation, from individual contributor to thought leadership.
• A serious tester who investigates products, not just verifies them. You have a considered position on how you know something is wrong and can teach it. (~8+ years of relevant experience typically produces this skill.)
• Fluent in the craft vocabulary and able to apply it under pressure: modeling, coverage, oracles, heuristics, chartering, note-taking, test framing, safety language, bug advocacy. (Rapid Software Testing, BBST, or equivalent grounding is a plus.)
• Risk-analysis depth: ability to produce plausible, specific, arguable failure stories and explain the reasoning so a team can reproduce it.
• Experience with decision logic and configuration-driven behavior: rules engines, eligibility or underwriting logic, pricing engines, multi-tenant configuration variation.
• Track record of finding what green suites miss and explaining how you got there.
• Experience building a capability in an organization that didn’t have one, leaving it functional after you moved on.
• Exceptional written communication. Your notes, bug reports, and testing stories must be clear, honest about uncertainty, and hard to argue with for engineers, executives, auditors, and regulators.
• Influence without formal authority, especially delivering unwelcome findings to proud people in a way that gets you invited back.
• Comfortable with agentic AI coding tools (Claude Code, Cursor, Copilot) as part of a daily workflow, and appropriately suspicious of their output.

Benefits

Bestow offers flexible remote/hybrid work, meaningful benefits, equity, and substantial growth opportunities.