Misprint: Building the Future of Collectibles Trading 🚀

Misprint is the first-ever transparent, real-time marketplace designed specifically for trading card enthusiasts. Imagine Robinhood meets Pokémon cards—instant, clear, and powered by advanced machine learning.

We’re solving the biggest pain in collectibles today: inefficient, opaque pricing and slow transactions. At Misprint, our tech enables collectors and investors to discover the true, real-time value of their cards, effortlessly list dozens of items with just a single photo, and trade confidently in a dynamic bid/ask marketplace.

Why Join Misprint?

• Market-Changing Impact: We're redefining a massive, fragmented collectibles market by bringing stock-market efficiency to Pokémon cards, sports cards, and beyond.
• Cutting-Edge Tech: Our pricing engine analyzes millions of data points, delivering unmatched accuracy and transparency. Say goodbye to outdated spreadsheets and biased data.
• Speed & Simplicity: Listing cards on Misprint is over 20x faster than competitors, transforming how collectors manage their inventory.

The Team Behind Misprint

• Eva Herget (CEO): A former Goldman Sachs analyst turned Pokémon finance YouTuber who scaled a card-selling side hustle into a $500K ARR business in just 3 months—she's all-in on shiny cardboard.
• Jon (CTO): Mathematician and ML expert who dropped out of his PhD to revolutionize asset pricing through sophisticated algorithms and machine learning.

We’re backed by Y Combinator, passionate about collectibles, and committed to transparency and speed.

Who We're Looking For

If you’re excited about collectibles, fintech, ML-driven products, or marketplace innovation—Misprint is the place to build your career. Join us in transforming how collectors around the globe value and trade their treasures.

Misprint is a peer-to-peer marketplace for graded and sealed Pokémon cards — think StockX, but built for liquidity, transparency, and speed in collectibles. We’re YC-backed, operating in NYC, and growing quickly.

We’re looking for a high-agency ML/Data Engineer to own our market data and pricing systems end-to-end: ingestion, normalization, storage, quality, and the logic/models that turn messy real-world data into reliable pricing and product experiences.

This role sits at the intersection of data + backend engineering. You’ll build and operate the pipelines and services behind pricing and market data, and ship the backend work needed to keep those systems fast, reliable, and scalable in production.

We’re building the data layer that makes Pokémon cards feel tradable like stocks: real-time market signals, smarter pricing, better search/discovery, portfolio analytics, and liquidity tools that help collectors actually execute. If you like turning messy real-world data into features users can feel — this role has a ton of surface area, and is core to our business.

What you’ll do

You’ll work closely with the engineers to:

• Own market data pipelines (scrapers/APIs/ETL) and make them reliable, monitored, and well-documented
• Improve pricing + liquidity logic (cleaning, normalization, deduping, outlier handling, stale-data handling)
• Build systems for data quality + observability (alerts, retries, backfills, audits, freshness metrics)
• Own core datasets/mappings that power product (catalog enrichment, ID mappings, pricing tables)
• Build and maintain backend surfaces that serve data to the product (APIs, queries, internal tooling)
• Partner with product/engineering to ship data-powered features (market stats, trend signals, portfolio analytics, search relevance)

What success looks like (first 60–90 days)

• Data pipelines run reliably with monitoring + alerting and clear recovery paths when things break
• “Critical scripts” are in the repo, reproducible, documented, and not tribal knowledge
• Pricing data is measurably cleaner, fresher, and more trustworthy with clear freshness metrics
• You can confidently ship improvements that increase pricing accuracy and marketplace liquidity

Our stack

• Backend/Data: Postgres (Supabase), Python, some Node
• Infra: Vercel, AWS, Porter, Stripe
• Frontend: React / Next.js / Tailwind (not the focus)

You might be a fit if you…

• Have 3+ years experience in data engineering / ML engineering / backend data systems
• Have owned production pipelines end-to-end: ingestion → transforms → storage → serving
• Are strong at Python + SQL, and comfortable with Postgres in production
• Know how to handle real-world data messiness: duplicates, schema drift, missingness, outliers, backfills
• Can operate independently, communicate clearly, and document systems as you go
• Prefer pragmatic systems that stay reliable over time

Bonus: scraping/automation experience, marketplaces/fintech data, search/relevance, time-series/anomaly detection, or you like Pokémon/collectibles

How to apply

Send:

• Your LinkedIn/GitHub if not in your Work at a Startup Profile (or anything you’re proud of)
• A short note on why you’re interested and what you’ve owned end-to-end
• (Optional) Any experience with Pokémon/TCG/collectibles!

Skills

Python, SQL, PostgreSQL, Data Engineering, ETL, Web Scraping, AWS, Supabase, Monitoring/Alerting

Misprint Tech Stack & Challenges 🛠️

At Misprint, we're leveraging a modern, robust tech stack (Next.js, Supabase, AWS, Python) to tackle complex challenges in the collectibles marketplace.

Technical Challenges We're Solving:

• Real-Time Data Processing: Building and optimizing systems that handle massive data volumes for instant pricing insights and trade execution.
• Machine Learning at Scale: Creating ML models sophisticated enough to accurately price hundreds of thousands of unique collectibles, continuously updating based on live market data.
• Seamless UX/UI: Designing workflows that allow collectors to effortlessly list and manage large inventories, improving marketplace efficiency by over 20x compared to competitors.