{"id":"2076571299434905692","url":"https://x.com/sammy_capital/status/2076571299434905692","text":"","author":{"name":"Sammy","username":"sammy_capital","avatarUrl":"https://pbs.twimg.com/profile_images/2072780575857860608/AXyNXyUB_200x200.jpg"},"createdAt":"Mon Jul 13 07:35:53 +0000 2026","engagement":{"replies":3,"retweets":20,"likes":137,"views":396269},"article":{"title":"How You Can Break Into Quant Finance From Zero in 16 Weeks","previewText":"Quants sit at the intersection of mathematics, software, and markets.\nThey build models that price risk, detect patterns inside enormous datasets, and execute decisions at a speed no human trader","coverImageUrl":"https://pbs.twimg.com/media/HNBhzksW0AAOH2i.jpg","content":"Quants sit at the intersection of mathematics, software, and markets.\n\nThey build models that price risk, detect patterns inside enormous datasets, and execute decisions at a speed no human trader could match manually.\n\nThat combination is why the best firms pay so much for them.\n\n![](https://pbs.twimg.com/media/HNBNP1xW8AE8Uoh.png)\n\nA strong quant can improve how billions of dollars are priced, traded, or protected. Even a small edge becomes extremely valuable when it is applied across enough capital and enough trades.\n\nBut getting into the industry is intentionally difficult.\n\nThe information is scattered across university courses, interview books, coding platforms, research papers, Reddit threads, and vague advice from people already inside. \n\nMost candidates spend months learning impressive material without knowing whether it will ever appear in the job or interview they want.\n\nBefore you keep reading:\n\nBookmark this guide so you can return to the 16-week plan.\n\n> And follow @sammy_capital  I break down quant finance, market models, and the decision-making systems used by people who trade and manage serious capital.\n\nI spent weeks reverse-engineering this roadmap from conversations with more than 30 quants.\n\nEverything below is the version I wish someone had handed me on day one.\n\n## Step 1 - Build the mathematical floor\n\nMost people begin quant prep by opening the hardest book they can find.\n\nThat usually wastes time.\n\nYour first goal is to build a mathematical floor strong enough that interview questions stop feeling like tricks.\n\n![](https://pbs.twimg.com/media/HNBWgYVWUAA-PKO.jpg)\n\nStart with:\n\n- probability;\n\n- expected value;\n\n- conditional probability;\n\n- variance and covariance;\n\n- combinatorics;\n\n- basic calculus;\n\n- linear algebra;\n\n- statistics.\n\nThe most important idea is expected value:\n\n![](https://pbs.twimg.com/media/HNBORUXWUAE8GKX.png)\n\nIf a game pays $120 with probability 0.40.40.4 and loses $50 with probability 0.6:\n\n![](https://pbs.twimg.com/media/HNBPwXRXYAAAi_W.png)\n\nSo the game is worth $18 per attempt before costs.\n\nQuant interviews then make it harder by changing the payout, hiding information, or asking how much you should pay to play.\n\nNext comes conditional probability:\n\n![](https://pbs.twimg.com/media/HNBP2PkWEAAIBO9.png)\n\nAnd Bayes’ theorem:\n\n![](https://pbs.twimg.com/media/HNBP42dXMAEfQ80.png)\n\nThen variance:\n\n![](https://pbs.twimg.com/media/HNBP7BZWoAADQ4B.png)\n\nExpected value tells you what happens on average.\n\nVariance tells you how unstable that average is.\n\nFor portfolio thinking, learn covariance and correlation:\n\n![](https://pbs.twimg.com/media/HNBP-aeXkAAOIWd.png)\n\nThis is where quant thinking starts to look different from retail investing.\n\nThe retail question is:\n\n> How many stocks do I own?\n\nThe quant question is:\n\n> How many independent risks do I actually own?\n\n## What to use\n\nStart with:\n\nThe Green Book by Xinfeng Zhou for probability, brainteasers, calculus, and linear algebra.\n\nThen use Heard on the Street for classic trading and research interview problems. These are the first two core books in the guide you shared.\n\nFor practice:\n\n- Green Book probability section;\n\n- Jane Street puzzles;\n\n- Jerry Qin’s probability list;\n\n- timed mock interviews;\n\n- Zetamac for mental math.\n\nFor traders, begin mental math immediately.\n\nUse:\n\n> [Zetamac](https://arithmetic.zetamac.com/) - aim for 40 first, then 50+\n[Math Trainer](https://mathtrainer.org/) — good for rebuilding speed from the ground up\n[TraderMaths](https://tradermaths.com/) — closer to real assessment formats\n[Wall Street Quants Mental Math ](https://www.thewallstreetquants.com/quant-mental-math-practice)— useful for “80 in 8” style drills\n\n![](https://pbs.twimg.com/media/HNBWvB7WwAA8bDO.jpg)\n\nStart around 20 on Zetamac and practice every day. A consistent 50–60 is usually a strong enough benchmark for many early trading screens.\n\nThe original roadmap suggests getting above 40 on Zetamac first, then pushing toward 50+ before interviews.\n\n> By the end of this stage, you should be able to solve expected-value problems, explain Bayes clearly, calculate basic variance and covariance, and talk through your reasoning under pressure.\n\nThat is the real foundation.\n\nBefore models, backtests, or stochastic calculus, you need to become comfortable making decisions under uncertainty.\n\n## Step 2 - Pick Your Track and Build Proof\n\nThis is where quant preparation splits.\n\nYou should not keep studying everything at once.\n\n- A Quant Trader is hired for fast reasoning under uncertainty.\n\n- A Quant Researcher is hired to find signals in noisy data.\n\n- A Quant Developer is hired to build the systems that make those ideas usable at scale.\n\nThat changes what you should practice.\n\n## Quant Trader\n\nFocus on:\n\n- probability and expected value;\n\n- mental math;\n\n- estimation;\n\n- market-making games;\n\n- fast verbal reasoning.\n\n![](https://pbs.twimg.com/media/HNBW5QTW8AA1ElT.jpg)\n\nRead or watch:\n\n- [Jane Street’s official interview guide](https://www.janestreet.com/join-jane-street/interviewing/?utm_source=chatgpt.com)\n\n- [MIT Introduction to Probability and Statistics](https://ocw.mit.edu/courses/18-05-introduction-to-probability-and-statistics-spring-2022/?utm_source=chatgpt.com)\n\n- The Green Book and Heard on the Street\n\n- Jane Street puzzles and timed market-making exercises\n\nYour benchmark is not whether you can solve a problem eventually.\n\nIt is whether you can solve it clearly while the assumptions keep changing.\n\nUse Zetamac, TraderMaths, the Green Book, and company-style probability questions.\n\n## Quant Researcher\n\nFocus on:\n\n- Python;\n\n- statistics and regression;\n\n- time series;\n\n- machine learning;\n\n- backtesting;\n\n- market microstructure.\n\n![](https://pbs.twimg.com/media/HNBXUhRWkAAH-B9.jpg)\n\nRead or watch:\n\n- [MIT Analytics of Finance](https://ocw.mit.edu/courses/15-450-analytics-of-finance-fall-2010/?utm_source=chatgpt.com)\n\n- [MIT Mathematics with Applications in Finance](https://ocw.mit.edu/courses/18-642-topics-in-mathematics-with-applications-in-finance-fall-2024/?utm_source=chatgpt.com)\n\n- [Scikit-learn’s PCA guide](https://scikit-learn.org/stable/modules/decomposition.html?utm_source=chatgpt.com)\n\n- An Introduction to Statistical Learning\n\n- Algorithmic Trading by Ernie Chan\n\nA basic research model often starts here:\n\n![](https://pbs.twimg.com/media/HNBQyYQWYAALzO5.png)\n\nYour job is to separate the return explained by common factors from the residual that might contain actual alpha.\n\nBut a backtest is only useful after fees, slippage, and out-of-sample testing.\n\n## Quant Developer\n\nFocus on:\n\n- C++;\n\n- data structures and algorithms;\n\n- operating systems;\n\n- networking;\n\n- concurrency;\n\n- performance profiling.\n\n![](https://pbs.twimg.com/media/HNBXk1UXoAAyg6Y.jpg)\n\nRead or watch:\n\n- Building Low Latency Applications with C++\n\n- C++ Core Guidelines\n\n- Computer Systems: A Programmer’s Perspective\n\n- MIT or university lectures on operating systems and networks\n\n- Exchange documentation explaining order types and limit-order books\n\nFor low-latency roles, correct code is only the beginning.\n\nYou also need to understand where time disappears:\n\n![](https://pbs.twimg.com/media/HNBQ9sOXgAAbuDW.png)\n\n## Use the right platform for the right role\n\nMyntBit is the strongest all-in-one option from the guide.\n\nIt separates preparation into Developer, Researcher, and Trader tracks, with C++, Python, brainteasers, finance MCQs, trading games, and company-style questions associated with firms such as Jane Street, Citadel, and Two Sigma.\n\n![](https://pbs.twimg.com/media/HNBXxMiX0AAlY2L.jpg)\n\nThat matters because most candidates waste time solving problems that have nothing to do with the role they are targeting.\n\n[QuantQuestion](https://quantquestions.io/) is useful for finance-heavy preparation.\n\nIt has a large interview-question bank covering portfolio construction, risk, and market concepts that pure coding platforms often ignore.\n\n![](https://pbs.twimg.com/media/HNBX5lRWEAA4zsY.jpg)\n\n[Quantable](https://www.quantable.io/) is especially useful for Quant Trader preparation because of its probability problems, company-tagged questions, detailed solutions, and interactive games.\n\n[LeetCode](https://leetcode.com/) remains essential for Quant Developer roles.\n\n## Build one serious project\n\nDo not build ten shallow notebooks.\n\nBuild one project that makes your target role obvious.\n\n![](https://pbs.twimg.com/media/HNBYFG9XEAAwLJ9.jpg)\n\nFor example:\n\n> Trader - market-making simulator with inventory and risk limits\nResearcher - factor model, options pricer, or stat-arb backtest\nDeveloper - C++ matching engine or limit-order-book simulator\n\nYou should be able to explain every assumption, failure point, and trade-off.\n\nThe original guide separates preparation into the same three paths because each one rewards different skills and has a different interview process.\n\nBy the end of this level, a recruiter should be able to look at your work and immediately understand which quant seat you are preparing for.\n\n## Step 3 - Train for the Interview and Apply at Scale\n\nThe final stage is where studying turns into simulation.\n\nYou already know the concepts.\n\nNow you need to solve the right problems under time pressure and explain your reasoning clearly.\n\n## Quant Trader routine\n\nDaily:\n\n> 15–20 minutes of Zetamac\n5 probability questions\n1 market-making or estimation drill\n1 answer explained out loud\n\nUse:\n\n> Green Book probability section\nHeard on the Street\nJane Street Puzzles\nQuantQuestion\nQuantable\nTraderMaths\nMyntBit Trader track\n\nDo not only solve for the number.\n\nPractice explaining which assumptions matter and how your answer changes when the interviewer changes the setup.\n\n## Quant Research routine\n\nPractice:\n\n> probability and statistics\nPython coding\nregression and time series\nmodel validation\nbacktest critique\nresearch discussion\n\nEvery strategy should be reduced to:\n\n![](https://pbs.twimg.com/media/HNBTleEWgAAUoM9.png)\n\nThen ask:\n\n- Does it survive out of sample?\n\n- Does it survive another market regime?\n\n- Does it survive lower liquidity?\n\n- Does it survive realistic transaction costs?\n\nUse MyntBit’s Researcher track, QuantQuestion, Quantopian lectures, MIT OpenCourseWare, and An Introduction to Statistical Learning.\n\n## Quant Developer routine\n\nPractice:\n\n> LeetCode daily\nC++ internals\noperating systems\nnetworking\nconcurrency\nsystem design\n\nFor every solution, explain:\n\n![](https://pbs.twimg.com/media/HNBT0ayXYAAvwtQ.png)\n\nAlso study:\n\n> lock-free queues\nmemory pools\nTCP vs UDP\ncontext switching\ncache misses\norder-book updates\npartial fills\n\nLeetCode gets you through the generic coding screen.\n\nSystems knowledge is what makes you look like a serious low-latency candidate.\n\n## Run one mock interview every week\n\nUse a 45–60 minute format:\n\n> 10 minutes - mental math or probability\n20 minutes - role-specific technical questions\n15 minutes - project deep dive\n10 minutes - resume and motivation\n\nAfterward, classify every mistake:\n\n> knowledge gap\nspeed problem\nweak explanation\nbad assumption\npanic under pressure\n\nThen repeat the same type of problem within 48 hours.\n\nApply wider than the famous firms\n\nUse:\n\n> Northwestern FinTech’s Quant Internships GitHub repo\ncompany career pages\nLinkedIn alerts for “quantitative,” “quant developer,” and “quant researcher”\nQuantNet forums\nGlassdoor for interview patterns\n\nStart with the obvious firms:\n\n> Jane Street\nCitadel Securities\nTwo Sigma\nHudson River Trading\nDE Shaw\nRenaissance Technologies\n\nBut also apply to:\n\n> SIG\nIMC\nOptiver\nJump Trading\nDRW\nVirtu\nFive Rings\nAkuna Capital\nFlow Traders\n\nTrack every application in a spreadsheet:\n\n> firm\nrole\ndate applied\ncurrent stage\ninterview format\nquestions asked\nweakness exposed\nnext action\n\nThe goal is to create enough interview volume that each rejection gives you information instead of ending the process.\n\nBy week 16, you should have one target role, one defensible project, a daily interview routine, and a live application pipeline. \n\nThat is when you stop looking like someone who is interested in quant finance and start looking like someone who trained for it deliberately.\n\n## Final Note\n\n![](https://pbs.twimg.com/media/HNBVJ7IWoAAcITH.png)\n\nQuant finance rewards people who can think clearly under uncertainty, build systems from math, and explain their decisions when the answer is not obvious.\n\nYou do not need to master everything at once.\n\nPick one track, follow the roadmap, build something real, and let each interview show you what to improve next.\n\nIf you want to keep going, these three posts cover the ideas underneath quant thinking:\n\n1. [Game Theory and Unpredictable Strategy](https://x.com/sammy_capital/status/2074215205643440384?s=20)\nWhy the strongest strategy is sometimes to remove your own pattern entirely.\n\n2. [Portfolio Construction and the Mutual Fund Theorem](https://x.com/sammy_capital/status/2074791465163563478?s=20)\nWhy professional investors start with risk-adjusted portfolio design instead of individual stock picks.\n\n3. [Options as Implied Probabilities](https://x.com/sammy_capital/status/2075183567970124117?s=20)\nHow option prices encode the market’s expectations about future outcomes.\n\nRead those next, then come back to the roadmap and choose which part you want to go deeper on."}}