Market Data Engineer — Prediction Markets & Odds Infrastructure

**About Us** Betstamp is the pricing intelligence layer for sports betting. We turn 200+ sportsbooks, exchanges, and prediction markets into one sharp, real-time view of fair value — for the sportsbooks that price against it, the bettors who bet with it, and the traders who move on it. We launched in 2020, built by professional bettors who needed sharper tools than the market offered and won at scale using them. That same pricing now runs as infrastructure across the industry. Because it's still used to win real money, it can't drift — every model is backtested against years of closing lines and monitored continuously. What our customers run on is the pricing we stake our own edge on. **What We Build** One pricing engine, delivered three ways: - **Sportsbooks and operators** get fair-value pricing on player props, alternates, and ladders, correlation-aware same-game-parlay pricing, and a normalized real-time and historical feed that drops into their existing trading stack. - **Professional bettors** get Betstamp PRO, the professional odds screen — True Line pricing across 200+ books and real-time edge detection the moment a market is mispriced. - **Prediction market traders** get the raw feed by API: sub-second data on every market we cover, benchmarked against the full spread of books and exchanges. 200+ sportsbooks and operators. 2,500+ markets. Sub-second refresh. **Why Betstamp** We're at the point where a product built for bettors has become infrastructure the industry runs on — and prediction markets have opened a category that barely existed three years ago. That's a rare window, and we're small enough that the people who join now shape what gets built rather than maintain what already exists. The work has unusual clarity to it. Pricing is either sharp or it isn't, and our customers would know within a day if it weren't. You'll see the effect of your work quickly, and you'll be trusted with real scope early. **Careers** Learn more at **Why Betstamp** Prediction markets are the fastest-growing category in trading, and the market data layer for them barely exists — we're building it. This is market-data infrastructure work with the clarity of a trading desk: latency and accuracy are the product, the benchmark is public, and you'll know within a day if something drifts. **The Role** You'll work on the real-time market data platform: ingestion from 100+ sources, a Redis Streams and ClickHouse pipeline, and the API that delivers line movement, best price, edge, and expected value at sub-second refresh. Think tick data, but for odds — normalization, latency benchmarking against the fastest sources, and the reliability practices behind a feed people trade on. AI is central to how we work. You'll be expected to use coding agents and other AI tools heavily — while reviewing their output carefully and remaining accountable for the result. ### What You'll Own - **Feed latency:** Benchmark our data against the fastest sources in the market and close the gaps. - **Ingestion:** 100+ integrations across books, exchanges, and prediction markets — kept alive through feed changes, renamed markets, and blocking. - **Pipeline:** Redis Streams, the market processor, and ClickHouse schemas built for line-history queries at scale. - **Reliability:** Prometheus/Grafana monitoring that catches degradation before customers do. - **AI-enabled development:** Automate the repetitive work; spend your judgment on the hard parts. Architecture decisions, production incidents, and irreversible changes will be handled alongside a senior engineer. ### The Stack Go 1.25 · PostgreSQL 18, Atlas, and ClickHouse · Redis/Dragonfly, Redis Streams, and RabbitMQ · Kubernetes, Docker, and GitHub Actions ### What We're Looking For - Experience with real-time data, streaming systems, or market data — professionally or through serious personal projects - Strong fundamentals in a backend language (Go, Java, C#, Rust, C++, Kotlin, or server-side TypeScript) - You think in milliseconds and percentiles — latency work should genuinely appeal to you - Comfortable with SQL and query performance investigation - Active, thoughtful use of AI coding tools, with the judgment to verify their work - Based in Toronto or willing to relocate Go experience isn't required, but you should be able to become productive in it within a few weeks. ### Nice to Have - Experience scraping websites or reverse-engineering undocumented APIs - Experience with ClickHouse, Prometheus, or high-throughput streaming systems - Experience building internal tools with LLM APIs - An interest in sports, betting, or prediction markets ### What’s In It for You - Competitive salary & raises — share in the upside you build - Autonomy & influence — real scope early, not tickets - Gear & learning stipend, plus free rein testing and spending on the AI tools of the future - World-class betting tools — the same tools professional bettors pay for, free - Sports-centric culture — team watch parties, office prop pools, and the inside scoop on sharp bets ### The Process - 45-min intro call — meet the team, swap questions - Lightweight take-home (under 2 hours) — a real problem from this codebase, not a LeetCode exercise - Systems conversation — a real problem together, no whiteboard algorithms - Final interview — meet leadership - Offer → onboarding → shipping to thousands of bettors