The problem MakeBanc is solving
Trillions in stablecoins sit idle because the infrastructure to deploy them into real, professionally managed strategies, safely and transparently, without surrendering custody, doesn't exist yet.
The strategies do. The pipes don't. So we built them.
The problem you'd primarily focus on
MakeBanc lives at the seam between on-chain and off-chain finance. Users connect funds on-chain. Capital flows through institutional custody to exchange venues where external managers trade. State has to stay coherent across smart contracts, custodians, exchanges, and a user-facing app, with settlement windows, custody trust boundaries, and financial-grade safety rails in the middle.
This is harder than a pure DeFi protocol and harder than a pure fintech backend. It's the interface between them, and the next stage of scale puts the most interesting architectural decisions in your hands.
The failure modes we design against are real: vault drains, NAV oracle manipulation, custody compromise, settlement window slippage, attestation failures. A vault failure at meaningful AUM is largely unrecoverable. So we design for it, audit for it, peer-review for it, insure against it, and stay paranoid about it.
You will hold this entire platform in your head. Web2 and web3. Integration partner meetings and back-end code. NAV oracle math and how the Asset Manager perceives it. If that sentence excites you, keep reading. If it stresses you out, it should.
Who you'd be working with
An intentionally small team, fewer than 10 people, with hands-on experience building startups, fintechs, and quant strategies. All with deep crypto understanding and enough scars to show for it. A 20-year trader as Chief Investment Officer, a 5th time Founder as CEO and a Chief Product Officer that has been in Crypto since 2017 and built a home financing platform to $8bn in loans. A flat structure where every senior engineer owns a domain end-to-end with no middle management. No ticket queues. No status meetings.
The platform serves distinct user types, all of whom you'll be designing for:
- Investors, the people allocating into strategies. They want transparency, control and confidence.
- Asset Managers, the external trading firms running strategies. They want clean integration, accurate NAV, and predictable operations that give them control and clarity.
AI Agents are first-class team members on this team, not tools. They have names, scopes, OKRs, and health metrics. You'll need to build yours on day one. They show up in incident response, in code review, in factsheet production. We treat them as headcount-equivalent capability, not as a tab in your browser.
How we operate
- Reliability is a feature. We treat uptime, correctness, and observability as user-visible product attributes, not back-office concerns.
- Every meaningful decision gets written down. Architecture Decision Records, post-mortems, threat model updates. Auditable trails matter more than meetings.
- Peer review at every gate. Code, content, agent output, runbooks, contracts. Reviewers raise issues. Owners approve. No self-merging on anything that touches customer capital.
- Design for the wrench attack, the rug pull, and the boring outage equally. Catastrophic and chronic failures both demand explicit defense.
- Builders only, including at senior level. No oversight-only roles. Senior engineers stay on the tools. Architecture happens in code review, not on whiteboards.
What you'd actually do
In your first few months you'd likely:
- Expand and own the architectural shape and roadmap of the platform across web2 and web3, from web app to vault contracts
- Own the difficult parts end to end: the NAV oracle pipeline, the database security setup, the orchestration layer that routes work between named agents across the platform
- Be the senior technical voice in partner integration conversations with institutional custodians (Ceffu and others), exchange venues, and oracle providers
- Shape the APIs and data models that the user-facing app, custody integrations, and external managers all depend on
- Build the operational backbone: monitoring, runbooks, and the safety rails that let a small team run serious financial infrastructure without drama
- Own three agents end-to-end as named team members: vault telemetry, NAV oracle operations, and the agent orchestration layer itself
- Sit in the Incident Commander rotation alongside the rest of the team
- Push back when a shipping decision degrades the platform shape, even when pushing back is inconvenient
A good week looks like: two partner integration deep-dives, one architecture decision documented as an Architecture Decision Record (ADR), twelve to twenty pull requests reviewed, three pull requests of your own merged, one taskforce you're a member of (not running, just contributing to), thirty minutes with our Solidity engineer on an on-chain edge case, an hour with Operations on an incident tabletop.
Our stack
What you'd inherit on day one
- Platform architecture across web2 and web3, including custody integration, client database structure, NAV oracle pipeline, attestation, vault telemetry
- The agent orchestration layer that routes work between named agents
- Technical seat at every partner integration table
- Ownership of three named agents: vault telemetry, NAV oracle operations, agent orchestration
- Membership in the Incident Commander rotation alongside the Head of Operations and the CEO
First 30 days: ratify your Q1 objectives with the leadership team, accept agent ownership, identify your first cross-functional taskforce
You'd thrive here if
- You think architecturally. You design systems, not just features, and can defend your trade-offs without resorting to buzzwords
- You've been senior before. Not "I had a team doing that" senior. Senior in the sense of holding architectural responsibility for a production system that survived contact with reality
- You have real web2 and real web3 production experience. Not a Solidity tutorial, from Web2 consumer apps to effective SAAS solutions, you've built them in real life.
- You've been in institutional integration partner meetings before. You know what institutional clients care about, and can tease out the unspoken needs in partner meetings.
- You're fluent with data: modelling, reconciling, making it trustworthy enough to move money on
- You understand crypto properly: transactions, oracles, signing, EVM, contracts. You can read Solidity
- You're security-minded by default. "Move fast and break things" breaks people's money, and you know it. You've worked on strong security focussed infrastructure before.
- You use LLMs and agents aggressively and with judgement. Bring your own Clawbot. We care about secure, structured output, not lines typed. Agents are first-class team members here, with their own scope and accountability, not a tab in your browser
- You're comfortable saying "I was wrong six months ago, here's what I learned." If you can't say this in interview, you'll struggle here
Bonus points
Experience with: institutional custody integrations (Ceffu, Fireblocks, Copper, BitGo), AWS Nitro Enclaves or other trusted execution environment work, NAV oracle or pricing infrastructure, async deposit and redeem patterns (ERC-7540 or similar), CEX APIs at production scale, oracles (Chainlink, Pyth, custom), Solidity at hobby level or above, indexers and event-driven systems, multi-tenant architecture for institutional clients, wealth management or brokerage backends, agent orchestration frameworks (LangGraph, Modal, custom).
This isn't for you if
- You want a large team, a clear backlog, and quarterly planning
- You've moved up the ladder by getting further from the work
- You prefer specs handed down
- "Security-minded" sounds like friction to you
- "I had a team doing that" is a sentence you'd say in interview
- Agents on the team sounds like marketing language to you, not a real way to operate
What you get
- A seat at the architectural table from week one, not a year in
- The full surface of a serious crypto product: contracts, custody, exchanges, consumer, regulatory
- Agents on the team as named collaborators with their own scope, OKRs, and health metrics
- Competitive base plus meaningful equity. We benchmark to top of market for senior platform roles in APAC and EMEA. We make our best offer first.
- Remote, EMEA or APAC hours.
Hiring process
We respect your time. Here's the full process from application to offer:
| # | Step | What to expect | Timing |
|---|---|---|---|
| 1 | Application review | The team reads every senior application personally. You hear back either way. | Within 5 days |
| 2 | Founder conversation | 45 minutes with one of the Founders. What you've built, why this role, what you'd want from the team. Less an interview, more a mutual fit conversation. | Week 1 to 2 |
| 3 | Technical conversation | 60 minutes with an engineering peer. Deep dive on a system you've owned. We want to understand how you think, not catch you on a puzzle. | Week 2 |
| 4 | Case Study | A real integration prototype against a sample partner spec we provide. The test is the same week of work you'd be doing in the role. Refusing the test self-selects out. | Week 2 to 3 |
| 5 | Reference checks | Two structured reference conversations with people you've shipped with. We tell you who we plan to call before we call them. | Week 3 to 4 |
| 6 | Offer | We make our best offer first. We don't negotiate down. If something's off, we have a real conversation. | Week 4 |
How to apply
Apply online and share:
- A paragraph on why this, why now
- Something you've built, broken, or written about. GitHub, a deployed thing, a post-mortem, anything real
- One architectural decision you've made and would defend, plus one you'd undo if you could
We've built something that should exist and didn't. If that's a sentence that gets you out of bed, apply.
Apply Now