Senior Software Engineer
Straight from Alt’s careers page. Apply on the company site — no recruiter, no middleman.
Senior Software Engineer - Credit
Department: Product and Data
Location: Remote
Compensation: $200K – $220K • Offers Equity
Employment Type: FullTime
Alt is unlocking the value of alternative assets, starting with the $5 B trading-card market. We let collectors buy, sell, vault, and finance their cards in one place and we are backed by leaders at Stripe, Coinbase, Seven Seven Six, and pro athletes like Tom Brady and Giannis Antetokounmpo. Our next frontier is real-time pricing at scale—the Alt Value that powers every trade, loan, and product on the platform.
THE ROLE
A collector borrows against a card sitting in our vault. We size the advance off its Alt Value, underwrite it, disburse the money, and manage the loan until the last dollar comes back or the collateral is released. Today too much of that path runs on people: manual approvals, hand-checked disbursements, a triage queue that grows when a transfer goes sideways, and screens that dont always tell the borrower where their money is.
Your job is to turn that into a product. Youll implement our new underwriting design hand in hand with our data scientists, automate the loan workflow end to end, own disbursement failures until they stop happening, and ship the borrower and operator experiences on top, built from our designs. This is a full-stack seat: the same engineer who fixes the idempotency bug builds the screen that tells the borrower their funds are on the way.
The north star: T7 (trailing 7-day) Disbursement within SLA rate, with a target of 99%. Every underwriting change, automation, triage fix, and screen you ship either moves that number or protects it. Supporting KPIs: time from application to underwriting decision, share of loans that go from application to funded with no manual touch, disbursement exceptions per week, and support tickets asking where is my money, driven toward zero. In lending, a disbursement that probably went out is a failure.
The people who are great at this have owned a lending or payments product where it mattered, from the state machine to the pixels. They think in idempotency and reconciliation because theyve been on the wrong side of both, and they care that the UI tells the truth about what the system is doing.
WHAT YOULL OWN
The underwriting revamp. Were redesigning how we underwrite cash advances and term loans, and youll implement the new design end to end, working hand in hand with our data scientists. That means turning their models and credit policy into a production decisioning service: inputs computed reliably, every decision versioned and explainable, and every rule or model change shipped through shadow mode and backtests before it goes live. Done looks like the new logic deciding every application, with Data Science able to iterate on it without an engineering rewrite.
The automated loan workflow. Application, underwriting decision, offer, acceptance, disbursement, servicing, repayment, delinquency, and collateral release, modeled as an explicit state machine with every transition logged. Manual steps become rules, and rules become auditable. Done looks like a standard advance funding with no human in the loop, and a non-standard one landing in front of the right person with the context already attached.
Disbursement triage, fixed at the root. Failed, duplicated, delayed, and stuck disbursements. Youll find out why each class of break happens (a retry without an idempotency key, a webhook that arrived out of order, a rail that returned after we marked it settled) and close it for good. Done looks like the triage queue shrinking week over week, with the T7 SLA rate to prove it.
The borrower experience, built to design. The cash advance offer, acceptance, disbursement status, repayment, payoff, and loan history screens across web and mobile, implemented faithfully from our designs and improved through what you learn in production. You own design fidelity, accessibility, and performance, and you make sure the status a borrower sees is the status the loan is actually in.
Operator and triage tooling. The internal console where Credit, Operations, and Support see every loan and disbursement, spot the stuck ones, and resolve them with the context already on screen. Watch what they do in spreadsheets today, then delete the spreadsheet.
Loan servicing and money movement. Scheduled auto-pay, partial payments, payoff quotes, reversals, and the edge cases between them, over ACH, RTP, and wires. Every movement integrates cleanly with the ledger so it posts, reconciles, and has an owner. Youll work closely with the Ledger team and own the product consequences of settlement timing, not just the API call.
Risk controls in the workflow. Limits, holds, loan-to-value checks, and step-up review triggered by risk signals rather than blanket friction. Youll know what each control costs in conversion.
The bar. Rigorous code reviews, clear technical docs, and mentorship for the rest of the Credit team. Youll be the teams closest partner to Design and Data Science.
WHAT GREAT LOOKS LIKE IN 6 MONTHS
T7 Disbursement within SLA rate is at or trending to 99%, and every remaining class of miss has a known cause and an owner.
Standard advances fund end to end with no manual step, and you can show the automation rate climbing.
The redesigned borrower flow is live, matches the design spec, and Support answers where is my money from the screen instead of escalating to engineering.
Operations clears the daily exception queue in under an hour, in a console you helped build.
At least one thing you shipped runs with AI in the loop, and the team depends on it.
The new underwriting logic is live in production, built to the agreed design with Data Science, with every decision explainable and every rule or model change versioned and auditable.
HOW YOULL USE AI HERE
We dont want to hear that you use Claude every day. Everyone does. We want to know what youve built with it. Concretely, in this role:
Stand up an agent that triages the disbursement exception queue overnight, clusters the breaks, and hands you a ranked list with a proposed root cause for each instead of a wall of failed transfers
Build a rail simulator that produces the failures a sandbox wont: a late ACH return, a duplicate callback, a webhook that arrives twice out of order, a disbursement that settles after the loan was canceled. Then run the workflow against it
Use models to generate adversarial test cases for the loan state machine and find the transitions nobody specified
Go from a design file to a working, typed component scaffold with tests in an hour, then spend your time on the details that make it feel right
Put an MCP server over loan and disbursement history so what happened to this borrowers advance is a sentence, not another one-off query
Use Claude Code or Cursor as the default way you work through a migration or a refactor, not as autocomplete
Build a harness that replays historical applications through the old and new underwriting logic side by side and explains every decision that changed
And have a view on the other side of it: what should not be handed to a model in credit decisions and financial code, and why. We ask.
If youve built agents, wired up connectors, written skills or MCP servers, or shipped internal tooling your team actually depends on, lead with that. Name the artifact, not the tool.
WHAT YOU BRING
5+ years building and scaling production applications across the stack: backend (Python, Postgres, AWS or comparable cloud infrastructure) and frontend (React, React Native, TypeScript). Both halves are load-bearing here.
Youve built or owned a lending, credit, or payments workflow in production. Loan origination, servicing, or disbursement, and you can explain a design decision youd make differently now.
Youve moved real money on real rails. ACH origination and returns, RTP or FedNow, wires, or equivalent. You know what an R01 is and why return windows matter to a disbursement.
Workflow and state-machine design. You model long-running processes explicitly, with retries, timeouts, compensation, and a clear audit trail.
Operational debugging instinct. Youve taken over a noisy exception queue and made it quiet by fixing causes, not by adding alerts.
Youve implemented complex designs faithfully and quickly, and you can talk a designer through a trade-off without losing the intent.
Database mastery. Expert SQL and relational design, and transactions that hold under load during scheduled auto-pay batches.
API-native. Webhooks, idempotency, eventual consistency, and reconciliation are your default vocabulary.
Product-owner thinking. You understand the flow of funds and how to tell a worried borrower exactly where their money is.
Pragmatism. You know when a queue and a lambda are the answer and when they are three new partial-failure modes.
Youve implemented credit decisioning or underwriting logic in production alongside data scientists: turning a model or policy spec into a reliable, versioned, testable service, with shadow mode, backtesting, and a clean rollout.
Bonus: workflow engines (Temporal, Step Functions or similar); credit or risk modeling; direct experience with Stripe, Plaid, Modern Treasury, or Increase; asset-backed or secured lending; design systems or internal tooling for operations teams; KYC/KYB, bank partnerships, or audit-readiness exposure. You collect, or you have a view on where this market is going.
HOW WELL INTERVIEW YOU
Well be specific so you can prepare. Before any of it: download the app, put a card up, make a bid, and try selling. Every round assumes youve used the product, and we expect your point of view to sharpen as you go. For this role, go through the cash advance flow: how an offer is presented, what happens after you accept, what youre told while you wait for funds, and where it made you hesitate. We will ask.
Recruiter screen, 30 minutes, Google Meet. Come ready to talk about the metric you own today and how youre performing against it, why Alt specifically, and how youre actually using AI. Well align on comp and answer your questions.
Hiring manager, Mina Quon, 30 minutes
Culture interview, 30 minutes.
Technical Assignment - 45 minutes
Architecture and system design, 60 minutes.
Leore, our founder and CEO, 30 minutes. Everyone who joins Alt meets Leore. Come with a point of view on where youd take Alts credit product first.
Reference checks, 15 minutes with a former manager(s) .
Compensation: $200,000 – $220,000 per year, plus equity for all full-time employees.
What we offer:
For full-time Employees
We cover 85% of your medical, dental, and vision, and up to 50% for dependents. HSA/FSA available
Generous PTO that people actually take
Parental leave at full salary
$200/month wellness + $100/month home office stipend
Free DoorDash Pass membership, because dinner shouldnt be a decision
Remote-first (for most positions), with WeWork access when you want a room with other humans
401(k)
Alt Equity to all full-time employees
Similar remote jobs
All Software Engineer jobs →

Alt
Staff Quantitative Researcher, Pricing (Remote)


Discover More than 100,000 Hidden Remote Jobs Before Everyone Else
Unlock All Remote Jobs Today
Simple pricing. Big savings on Quarterly and Yearly.
Monthly Access
- Instant access to fresh remote jobs from 500+ companies
- New opportunities added hourly, often 3-7 days before anywhere else
- Advanced filtering by role type, stack, pay, and location
- Priority customer support
Yearly Access
- Everything in Monthly
- Save $169 (~74%) vs paying monthly
- Average job search takes ~6 months - get covered for the whole journey
- Less than the cost of one lunch per month for competitive advantage
- Equivalent to just ~$4.92/month