TetraFi
Senior Developer / Founding Engineer - Blockchain, AI, Trading Systems
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Senior Developer / Founding Engineer - TetraFi
Own execution end-to-end at the settlement layer institutions are moving onto.
About TetraFi
TetraFi is the multilateral settlement and liquidity rail institutions build on. One instruction: priced by competition, obligations netted, settled atomically across chains - compliance and audit trail built in. Live today: liquidity aggregation, firm-quote execution, escrowed atomic cross-chain settlement. Next: T+0 vaults, instant redemption, a programmable credit layer, multilateral netting.
The role
You are engineer #2 on a working platform; the team runs agent-first - coding agents are the primary development loop. Your mandate is the execution vertical: everything between a request for quote and funds settled across chains. You grow into owning it and run it like a product.
What you'll do
- Own the path from quote to settled funds: the RFQ engine, aggregator API, order lifecycle, solver and settlement contracts.
- Integrate liquidity - CEX and DEX venues, OTC desks, on-chain FX, bridges, payment and fiat rails - and decide which earn a place in the route.
- Push execution quality: tighter pricing, higher fill reliability, order tracing that lets an institution reconstruct any trade.
- Harden settlement across EVM, Solana, Tron and Cosmos: escrow, delivery-versus-payment, claims, failure paths.
- Run the vertical in production, incidents included - we run lean; you share the pager with the founding engineer.
Stack: Rust (axum, tokio) · Solidity + Foundry · Solana/Anchor · CosmWasm · cross-chain protocols (CCTP, LayerZero, Hyperlane) · Postgres, Redis · TypeScript/React · Claude Code/Codex agents, MCP
What we're looking for
Engineering
- Rust in production - required. You have built and operated async Rust (tokio) services others depended on - fluent in ownership and Send/Sync, cancellation safety, backpressure - with tracing-based observability, profiling, and allocation-aware hot paths as part of the work.
- Smart contracts - required. Production contracts that have held real value:
- Deep EVM fluency and custody-aware design - checks-effects-interactions, minimal external surface, escrow lifecycles end-to-end;
- Exploit-class fluency as a design input - access control, reentrancy, oracle and flash-loan manipulation, vault inflation;
- Fuzz and invariant testing as a default, fork tests, audits you've owned end-to-end - and live-contract discipline: multisig and timelock governance, pause paths, monitoring;
- Beyond the EVM - strong plus: Solana's account-model security (Anchor), CosmWasm's actor model.
- Execution systems - required. You've built the core of a trading system, in DeFi or against crypto venues:
- the order lifecycle from quote to on-chain fill - state machines, idempotent and recoverable, 24/7;
- routing across fragmented liquidity - CEXs, DEXs, RFQ - under fees, latency and fill probability, with inventory rebalanced across venues and chains;
- risk controls that fail closed - price, size and position limits, kill switches - and books-versus-chain reconciliation with live position and PnL.
- Cross-chain - strong plus. You understand cross-chain transfer models - burn-and-mint versus messaging - and what each trusts; you design for finality, reorgs and recovery; you know what "atomic" can and cannot mean across two chains.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
DeFi and finance
- Required - you can reason about this from first principles:
- Market structure - how liquidity is actually provided across RFQ, CLOBs and AMMs: firm versus indicative quotes and last look; spread as compensation for adverse selection and inventory risk; quote skew, flow toxicity and markouts; the winner's curse in multi-dealer RFQ.
- Execution quality - slippage versus arrival, market impact and how it scales with size, fill and reject rates, markout curves; best execution as an obligation - and its on-chain expressions: MEV, private orderflow, oracle-versus-venue pricing.
- Auction design - you understand auctions as market design: solver competition, Dutch decay, batch clearing; why failing to fill must cost something.
- Settlement and netting - who moves first, and why that is the risk: principal (Herstatt) risk versus credit risk; DvP and PvP versus the free-of-payment and on-credit settling crypto OTC still runs on; both sides of atomic-versus-netted settlement - atomicity kills principal risk, netting compresses obligations and frees capital, prefunding is the price; off-exchange settlement and collateral mirroring; finality as a probabilistic, per-chain threshold.
- DeFi protocols - you can explain how each is constructed and where it breaks:
- intents and solvers: ERC-7683 auctions (UniswapX, CoW, Across), signed-quote RFQ, and solver economics - fronting inventory, filling, then claiming repayment;
- AMMs: price impact, MEV, what LPs pay arbitrageurs - and why a pool price is never an oracle;
- lending: liquidation mechanics, cascades, bad debt - including today's modular, curated markets;
- perps: funding, mark versus index - where the liquidation price actually comes from - and the backstop layers behind it;
- vaults: share accounting and its attacks, and vault shares used as collateral;
- stablecoins: fiat-backed, CDP, delta-neutral - each with its own depeg path; secondary price versus redemption price;
- oracles: push versus pull, manipulation cost, and honest-but-divergent prices.
- Tokenization - you understand how tokenized products are actually built: transfer-agent mint and redeem, eligibility enforced in the token itself (allowlisted ERC-20s in practice, ERC-3643 as the named standard), oracle-published NAV and its staleness, instant versus queued redemption, rebasing versus price-accruing yield - and the mint/redeem arbitrage that keeps secondary price near NAV.
Get help with your application
Your very own career expert that helps elevate your application to the next level.
AI-native
- Required
- Agent-first development - spec-first, plan-mode workflows; rules files kept as a map, not a manual; skills and hooks maintained like code; closed verification loops (tests, browser, logs); fresh-context review of every agent diff.
- Agent orchestration - orchestrator–worker fan-out, worktree-parallel sessions, Ralph-style fresh-context loops with plan-files and verified completion, agent teams on shared task lists, CI-triggered and scheduled agents, agent review gating human review.
- Agent and AI infrastructure - multi-agent workflows and MCP servers (LangGraph, Claude/OpenAI agent SDKs); RAG and knowledge-graph retrieval (LlamaIndex, vector databases, GraphRAG); agent tracing and evals; model gateways and routing; sandboxed execution; self-hosted inference (vLLM-class).
You don't need a degree, or every part of our stack on day one. But if you want a ticket queue, a review layer above you, or AI means autocomplete, this isn't your role.
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location