About the role:
This is a hands-on technical leadership role, not a pure management one. You own the core
and infrastructure layer: the infrastructure, the ledger, and the settlement service — the parts
of the system where being wrong damages trust irreversibly. You also own the architecture of
the whole platform, the integration discipline that keeps parallel workstreams from diverging,
and the relationship with the client on scope and go/no-go decisions.
What you'll own:
• The money-grade foundations. A PostgreSQL double-entry ledger (serializable, multi-AZ,
RPO 0) where balances are derived from postings rather than stored as mutable counters,
with idempotency keys on every mutation and a reconciliation job that pages on any
discrepancy. When real money arrives, it must be new posting types — not a new system.
• Determinism and recoverability. A single sequencer per game, an append-only event
log, and snapshot-plus-replay recovery that produces bit-identical state. This is also what
makes the engine testable against the source simulations.
• Fail-safe behaviour, everywhere. The pricing workbook's no-arbitrage checks (sum of
YES asks > 1, sum of YES bids < 1, mids sum to exactly 1, YES/NO complements per
outcome) run as runtime assertions that halt the market and page on-call when
violated. A stale feed pauses the market; an unreachable engine produces a clean order
rejection. The system never fails open.
• The settlement service. Mid-game TIE/GTL resolution with correct no-intermediate-tie
sequencing, final-buzzer KTL resolution, a maturity fee on profit, and finalisation only after
reconciliation against official scoring — plus an admin manual-override and market-void
path for overtime, voided games, and league scoring corrections.
• Infrastructure as code. AWS EKS, RDS multi-AZ, Terraform; autoscaling for bursty,
schedule-driven game-night traffic; managed failover primitives. The engine is single-
threaded by design (determinism), so it scales horizontally by game — each game's
markets are independent — rather than by threads.
• Risk as a first-class operational signal. The maker's live exposure — the L2 norm of its
[KTL, TIE, GTL] position vector, residual absorption (U), distance to the throttle cap, and
cancel rate — is monitored and paged like infrastructure health.
• Delivery leadership. Hard integration checkpoints so workstreams never diverge, a pre-
agreed descope ladder that flexes scope while protecting the ledger / determinism / fail-
safe guarantees, the failure drills, and the launch-readiness review with the client.
Requirements
• Senior, hands-on experience building financial, exchange, trading, or ledgered systems
where correctness and auditability are first-class — payments, brokerage,
betting/wagering, or order-matching backgrounds all fit.
• Deep practical command of double-entry accounting systems and the discipline of
deriving balances from an immutable record.
• Real experience with event sourcing / CQRS: sequencers, append-only logs, snapshot-
and-replay, and why determinism matters for both recovery and testing.
• Strong PostgreSQL depth — serializable isolation, replication, recovery objectives — and
production ownership of AWS (EKS/Kubernetes, RDS), Terraform, multi-AZ failover.
• TypeScript / Node.js fluency; the stack is TS end-to-end and the engine spec translates
directly from the client's existing JS/TS code, so discrepancies surface fast.
• A track record of leading small senior teams to a fixed external commitment and
making the scope trade-offs that protect the core when something slips.
• Calm, direct client communication — you'll run the scoping sessions (settlement edge
cases, fee schedule, data-vendor selection) and the go/no-go review.
Nice to have:
• Prior work with a real-time matching engine or order book — ideally one with a
maker/inventory-risk model (exposure caps, skew, unwinding).
• Experience integrating a licensed sports-data feed (Sportradar / Genius Sports) or similar
low-latency authoritative source; scraping is disqualifying for a money product.
• Observability ownership at the SLO level (Datadog / PagerDuty / Sentry), including treating
business-risk metrics as paged, first-class signals.
• Familiarity with regulated/licensing-gated product paths (KYC/AML, responsible gaming,
SOC 2) — relevant to the real-money roadmap that this architecture is built to absorb.
Stack:
TypeScript / Node.js · PostgreSQL (multi-AZ) · NATS JetStream · Redis · WebSockets · React /
Next.js · AWS EKS / RDS · Terraform · Datadog · PagerDuty · Sentry
Benefits
At Remotebase, we are on a mission to bring together great ideas and great people, transcending physical borders. As the world embraces remote work, we stand at the forefront of this transformative shift, empowering exceptional companies to collaborate with top talent on a global scale.
Our purpose is clear: to help organizations build the best remote engineering teams comprising of the top 1% talent within 24 hours. We specialize in partnering with forward-thinking businesses that are actively shaping the world we inhabit. By deeply understanding their ideas, needs, and aspirations, we create innovative technology solutions that drive their success.
At the heart of our approach is the relentless pursuit of excellence. We meticulously source the finest TECH talent, curating passionate teams eager to tackle complex challenges head-on. Whether you require expertise in Software Engineering, UX Design, Product Management, Data Engineering, Data Analysis, Data Science, AI, QA, or DevOps, we have you covered.
What sets us apart is our unwavering commitment to delivering exceptional results swiftly. We understand the urgency inherent in today's fast-paced business landscape and are dedicated to providing you with a highly skilled, motivated team from day one. With Remotebase, your organization can hit the ground running, accelerate growth, and achieve remarkable outcomes.
Join us as we revolutionize the way companies thrive in a borderless world. Let's create a future where remarkable ideas and exceptional talent know no bounds.
