About Cache

Cache Financials is a bespoke brokerage offering investment products tailored towards individuals with large concentrated stock positions. Since launching in March 2024, Cache has scaled to over $2 billion in assets and is operating profitably.

We are a small, ambitious team that values ownership, transparency, customer focus, and sustainable speed. Reliability, security, and safety are non-negotiable.

We work fully remotely. The team meets in person once a week in San Francisco or New York.

The role

We’re looking for a Staff Backend Engineer to own key components of our infrastructure. You will design and evolve the services and data models behind account operations, investment workflows, money and securities movement, and external financial integrations.

You’ll work closely with product, operations, compliance, and other engineers to turn complex financial processes into systems that are understandable, observable, auditable, and safe to operate.

Backend services are written in Kotlin on Spring Boot and built with Bazel. Temporal orchestrates our long-running financial workflows, Protobuf is the contract layer with the frontend, and Aurora PostgreSQL is the system of record. We run on AWS with Pulumi, and monitor with Datadog.

Strong programming skills in a backend language are required. Experience with Kotlin or Temporal is helpful, but backend judgment matters more than prior experience with our exact stack.

Our primary hiring bar is not familiarity with a particular framework or even prior experience in investment platforms. It is your ability to reason precisely about distributed systems under uncertainty, identify the invariants that matter, and translate that reasoning into pragmatic systems a growing team can maintain.

What you’ll do

• Own the architecture and delivery of financial workflows spanning Cache, customers, and external partners.
• Establish clear system boundaries, sources of truth, invariants, and failure behavior before choosing implementation mechanisms.
• Model long-running workflows with explicit states and a clear distinction between requested actions and confirmed facts.
• Design APIs and event processing that behave safely under retries, concurrent requests, duplicate delivery, out-of-order messages, and ambiguous partner responses.
• Build durable processing patterns using transactions, queues, outboxes, idempotency, deduplication, reservations, and explicit recovery paths where appropriate.
• Design financial records so cash movements, positions, executions, and other economic effects are auditable and cannot be silently applied twice.
• Make reconciliation and discrepancy resolution first-class system capabilities rather than after-the-fact operational patches.
• Improve observability, operational tooling, runbooks, alerts, and safe controls for detecting, containing, and recovering from failures.
• Write clear design documents, lead architecture reviews, and help the team make better technical tradeoffs.
• Ship incrementally while balancing customer impact, operational risk, maintainability, and delivery speed.

What we’re looking for

• Significant backend engineering experience, including technical leadership over systems with meaningful correctness or reliability requirements. This often comes from 8+ years of experience, but scope and judgment matter more than tenure.
• Experience designing, building, and operating distributed systems involving databases, asynchronous processing, and external services.
• The ability to reason precisely about timeouts, retries, races, partial completion, duplicate or out-of-order events, and crashes between side effects.
• Strong data-modeling and transactional reasoning, including an understanding of where database guarantees end and application-level guarantees begin.
• Experience with systems that move money, or an equivalent domain where a silent double-spend is unacceptable.
• Experience building systems that are observable, auditable, recoverable, and operable by people other than their original authors.
• Strong written and verbal communication. You make assumptions, invariants, tradeoffs, and failure behavior legible to engineers and non-engineering partners.
• A track record of raising the quality of the team around you through technical direction, mentoring, design reviews, and reusable systems.
• High ownership and customer empathy, especially when failures can affect someone’s money or trust.
• Comfort moving quickly in an early-stage environment without removing the safeguards that correctness-critical systems require.

Nice to have

• Experience in brokerage, payments, custody, clearing, accounting, banking, logistics, or another correctness-sensitive domain.
• Familiarity with immutable ledgers, reservations, reconciliation, settlement, or financial audit trails.
• Experience with workflow orchestration, event-driven architectures, or external partners with imperfect reliability characteristics.
• Experience introducing property-based testing, fault injection, replay tooling, or invariant validation.
• Experience helping an early-stage engineering organization evolve its architecture and operating practices as the product scales.

Financial-domain experience is useful but not required. We value engineers who can learn a domain and derive correct designs from first principles.

What success looks like

In your first 30 days, you will understand Cache’s critical workflows, their sources of truth, and their most important failure modes. You will build trust across engineering, product, and operations and ship a meaningful improvement to a critical system.

Within three months, you will have led a substantial reliability improvement to reconciliation across our financial systems, and shipped tooling that lets operations diagnose reconciliation breaks without engineering involvement.