Help us build the
future of money

Full-time · Remote — EU / UTC-2 to UTC+4

You will own the code that moves money. The ledger is the single source of truth for every balance on the platform: deposits, tiered yield accrual, referral commissions and withdrawals all settle through it. It maintains a hard invariant — the sum of all active stake balances must equal the user balance, to the satoshi, at every instant — and your job is to keep that true while the system grows. This is not a CRUD role. You will design locking strategies for concurrent balance mutation, reason about isolation levels under real contention, and write migrations that run against a live production database holding customer funds.

Requirements • 6+ years building production backend systems in Python, with at least 3 years on services where a bug has direct financial consequence — payments, exchange, brokerage, lending or core banking. • Deep, demonstrable command of PostgreSQL beyond ORM usage: you can explain the difference between REPEATABLE READ and SERIALIZABLE in terms of the anomalies each permits, diagnose a deadlock from pg_locks, reason about MVCC bloat, and justify a chosen lock ordering as a proof rather than a convention. • Fluency in async Python (asyncio, SQLAlchemy 2.0 async ORM, FastAPI) including its failure modes: you know why a blocking call inside a coroutine is an outage, and how connection pool exhaustion presents itself in production. • Rigorous discipline around decimal arithmetic. You must be able to articulate, without prompting, why floating point is disqualified from any balance path, what ROUND_DOWN protects against, and where rounding dust accumulates in a proportional-reduction algorithm. • Experience writing and reverting schema migrations against live systems with zero-downtime constraints, including backfills of financial data where a mistake is not recoverable by re-running. • A track record of designing for idempotency: scheduled jobs that may fire twice, webhooks that may be replayed, and multi-worker deployments where the same code path runs concurrently in separate processes. • Ability to write clear technical prose. Every non-obvious decision in our codebase is documented at the point of the decision, and you will be held to that standard in review. Strongly preferred • Prior work on a custodial crypto system: deposit detection, confirmation thresholds, chain reorganisation handling, hot/cold wallet separation. • Experience with double-entry accounting models and reconciliation tooling that proves ledger integrity continuously rather than on audit. • Familiarity with formal or property-based testing (Hypothesis, model checking) applied to state machines that hold value. Selection process Four stages: a written systems-design exercise on a concurrent balance-mutation problem, a live debugging session against a seeded database with a real invariant violation, an architecture interview on failure modes, and a reference check. Expect roughly eight hours total. We do not use algorithmic puzzle interviews.

$110,000 — $165,000

Full-time · Remote — EU / UTC-2 to UTC+4

You are the adversary we keep on payroll. The platform holds customer funds behind a public API, a Next.js frontend and an admin surface, and every one of those is continuously probed. You will run offensive assessments against our own systems, drive the remediation of what you find, and build the controls that make the next finding harder. You will also own the incident path: when something does happen, you are the person who establishes what occurred, what was reachable, and what must be disclosed.

Requirements • 5+ years in application security, penetration testing or offensive security engineering, with at least 2 years assessing systems that custody funds or personally identifiable financial data. • Demonstrated ability to find authorisation flaws by reading code, not only by fuzzing: broken object-level authorisation, confused deputy patterns, privilege boundaries that hold in the handler but not in the service beneath it. • Working knowledge of the cryptography we actually depend on — bcrypt work factors, JWT algorithm confusion and the reason algorithms must be pinned on decode, TOTP replay windows, constant-time comparison — and the judgement to know which weaknesses are exploitable in context and which are findings on paper. • Practical experience hardening a real deployment: reverse proxy configuration, rate limiting that survives being behind a CDN, Content Security Policy tightened past the point where it breaks things, secret management, and least-privilege database roles. • Ability to reason about the trust boundary between a server-rendered frontend and an API: where a secret may live, what an httpOnly cookie does and does not protect against, and why an admin key that reaches the browser is a finding regardless of how it got there. • Experience running an incident end to end — containment, forensic timeline, root cause, customer communication — and the temperament to do it calmly at 3am. • Written communication strong enough that a report of yours is actionable by an engineer who was not in the room. Strongly preferred • Published CVEs, bug bounty track record, or CTF results at a level you are willing to have examined. • Threat modelling experience against custodial crypto systems specifically: withdrawal approval flows, address substitution, insider risk on the admin surface. • OSCP, OSWE, GWAPT or equivalent — useful as a signal, never a substitute for the work sample. Selection process A time-boxed assessment of a deliberately vulnerable build of our own stack, a written report graded on accuracy and exploitability rather than volume, a defensive-design interview, and a reference check.

$100,000 — $150,000

Full-time · Remote — EU / UTC-2 to UTC+4

The tiered return schedule the platform offers is a liability curve. Your work is to make sure the asset side sustains it. You will model the yield the strategy actually produces across market regimes, quantify the gap between promised and realised return under stress, and turn that analysis into concrete recommendations on tier structure, capacity limits and reserve policy. You will be expected to argue against the business when the numbers do not support it, and to show your work when you do.

Requirements • Advanced degree in a quantitative discipline — mathematics, statistics, physics, financial engineering or equivalent — or a research record that plainly substitutes for one. • 4+ years of applied quantitative research where your models informed decisions involving real capital, with references who can speak to how those models performed out of sample. • Rigorous grounding in time-series econometrics and stochastic processes, and the statistical honesty to distinguish a genuine effect from one manufactured by multiple testing, survivorship bias or a look-ahead leak in the backtest. • Expert Python for research: NumPy, pandas, SciPy, statsmodels, with the engineering discipline to produce work that a second person can reproduce from the repository alone. • Practical understanding of digital-asset market structure: liquidity fragmentation across venues, funding rate dynamics, impermanent loss, smart-contract and bridge risk, and the way correlations converge to one precisely when diversification is needed. • Demonstrated capacity to size and stress-test a liability schedule — to answer, with a defensible number, what redemption rate under what drawdown breaks the model. • Willingness to state an uncomfortable conclusion in writing and defend it. Strongly preferred • Experience at a fund, market maker, or treasury desk with direct responsibility for a book. • Familiarity with on-chain data pipelines and the pitfalls of treating on-chain activity as a clean dataset. • Prior work on capital adequacy or reserve modelling in a regulated context. Selection process A take-home research problem using a dataset we supply — you will present findings and be questioned adversarially on methodology — followed by a modelling interview and a reference check. We care about the reasoning, not the sign of the result.

$120,000 — $180,000

Full-time · Remote — EU / UTC-2 to UTC+4

The platform runs on infrastructure shared with other production services, which means every change you make has a blast radius beyond your own system. You will own availability, deployment safety, observability and disaster recovery. The bar is specific: a failed deploy must not take the database with it, a migration must be revertible, and we must be able to prove a backup restores — not assume it.

Requirements • 5+ years in SRE, platform or infrastructure engineering, including on-call responsibility for a system where downtime had a direct financial cost. • Deep Linux systems knowledge: systemd unit design including resource limits and restart semantics, journald, cgroups, network namespaces, and the ability to diagnose a host from first principles when the dashboards are the thing that is broken. • Production PostgreSQL operations: replication, point-in-time recovery, connection pooling, vacuum and bloat management, and tested restore procedures. You must have personally restored a production database from backup, and be able to describe how long it took. • Reverse proxy and TLS operations at a level that includes validating configuration before reload, reasoning about certificate renewal failure modes, and configuring rate limiting and header forwarding correctly enough that per-IP controls downstream still function. • CI/CD design where the pipeline is itself treated as production: deployments that health-check and roll back, migrations gated behind an explicit step, and secrets that never enter a build log. • Observability built for diagnosis rather than decoration — you instrument to answer questions you have not thought of yet, and you can articulate the difference between a health check that returns 200 and one that is actually meaningful. • Operational conservatism on shared infrastructure: you back up before you edit, you validate before you reload, and you never restart what a reload would have fixed. Strongly preferred • Experience running stateful services without container orchestration, where the host is the unit of deployment. • Background in security operations: host hardening, intrusion detection, log integrity. • Experience writing runbooks that a colleague can execute under pressure without you. Selection process A practical exercise on a deliberately broken environment, an incident-review interview covering a real outage from your past including what you got wrong, a design discussion on deployment safety, and a reference check.

$95,000 — $145,000

Full-time · Remote — EU / UTC-2 to UTC+4

The interface is where trust is either earned or lost. You will own the marketing site and the authenticated dashboard: a Next.js App Router application in nine languages, one of them right-to-left, rendered mostly on the server and held to a strict Content Security Policy. The work is equal parts product craft and rigour — a mislabelled balance or an ambiguous confirmation dialog is a financial defect, not a cosmetic one.

Requirements • 5+ years building production frontends, including at least 2 years on React with a modern server-rendering framework, and direct experience with the App Router model: server components, server actions, streaming, and the boundary between server and client. • Rigorous TypeScript. You use the type system to make invalid states unrepresentable, and you do not reach for `any` to end an argument with the compiler. • Genuine accessibility competence, not a checklist: semantic landmark structure, correct heading hierarchy, labelled and programmatically associated form controls, focus management in dialogs and route transitions, and testing with an actual screen reader. • Command of modern CSS and design tokens, including colour spaces beyond sRGB and the practical implications of oklch for perceptually-even palettes. You can measure a contrast ratio and defend it against WCAG AA rather than assert it. • Internationalisation experience covering more than string substitution: locale-aware number and date formatting, bidirectional text and logical CSS properties, and the discipline required when a missing key in one locale renders as nothing at all. • Demonstrated ability to hit Core Web Vitals targets on a real product and to explain each optimisation in terms of what the user perceives. • Structured-data and crawlability literacy, including how AI-driven retrieval reads a page differently from a traditional crawler, and what markup makes the difference. Strongly preferred • Experience with fintech or crypto interfaces where numeric precision, irreversible actions and error states carry real consequence. • Familiarity with strict CSP in a framework that would prefer you did not have one. • A portfolio of interfaces you designed as well as built. Selection process A take-home implementation of a real screen from our product, reviewed for accessibility and type safety as much as for appearance, followed by a code-review interview on your submission and a reference check.

$90,000 — $135,000

Full-time · Remote — EU / UTC-2 to UTC+4

You will build the compliance function rather than inherit one. That means establishing the onboarding controls, transaction monitoring, sanctions screening and escalation paths that a custodial platform requires, documenting them so they survive an examination, and advising the business honestly on where its current practice does not meet the standard it claims. You will report directly to the founder and you will be expected to say no in writing when that is the correct answer.

Requirements • 5+ years in compliance, financial crime prevention or risk at a regulated financial institution, payments provider or virtual asset service provider, with at least 2 years in digital assets. • Working command of the applicable regimes — AML/CFT obligations, the FATF Travel Rule, MiCA, GDPR — and, critically, the judgement to apply them to a specific business model rather than recite them. • Hands-on experience designing and operating KYC/KYB onboarding, including risk-based due diligence tiers, PEP and sanctions screening, and the false-positive burden each choice creates. • Practical transaction-monitoring experience: writing and tuning detection rules, investigating alerts, and producing suspicious activity reports that withstand scrutiny. • Familiarity with blockchain analytics tooling and the limits of attribution — you understand what a chain-analysis label is and is not evidence of. • Ability to write policy that engineers can implement and auditors can follow, and to maintain the evidence trail that demonstrates the policy was actually operating. • Demonstrated independence. We are specifically looking for someone who has previously escalated against commercial pressure and can describe what happened. Strongly preferred • Direct experience of a regulatory examination or licence application in an EU jurisdiction. • CAMS, ICA or equivalent certification. • Legal qualification, or experience working closely with external counsel on product structuring. Selection process A written case study assessing a realistic onboarding and monitoring scenario, an interview on regulatory reasoning and judgement under commercial pressure, and a reference check. Candour about the current state of our controls is expected in both directions.

$85,000 — $130,000