https://hasanjaved.me/blog/ Long-form essays on mathematics, finance, and AI by Hasan Javed — senior full-stack & AI engineer. en-us hasanjaved065@gmail.com (Hasan Javed) hasanjaved065@gmail.com (Hasan Javed) Wed, 22 Apr 2026 00:00:00 GMT scripts/generate-rss.mjs https://hasanjaved.me/blog/clash-of-titans-2026-ai-ipo-year/ https://hasanjaved.me/blog/clash-of-titans-2026-ai-ipo-year/ Wed, 22 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance Markets AI OpenAI Anthropic xAI SpaceX Wall Street Strategy Essay OpenAI, Anthropic, and SpaceX-via-xAI are about to drag $240 billion of public-market capital into one calendar window. The largest concentrated tech IPO season in history, and the moment the AI race stops being a private-market story and becomes a Wall Street one. — Three S-1s. Ninety days. $240 billion of fresh equity. Anthropic in October. OpenAI in Q4. SpaceX, carrying xAI inside it like a Trojan horse with a chatbot, in summer. There has never been a quarter like this in modern equity-markets history — and inside it, three radically different theories of what an AI company is will collide. The trial that could detonate it all begins in nine days. The strangest underwriter covenant ever printed is already on the syndicate desks. By 2028, two of the three listings will look obviously correctly valued. One will look catastrophically wrong, and the bear case will be the most-cited finance textbook of the decade. This is the dispatch from the cliff edge. https://hasanjaved.me/blog/algo-trading-and-machine-learning-shared-mathematics/ https://hasanjaved.me/blog/algo-trading-and-machine-learning-shared-mathematics/ Tue, 21 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) AI Algo Trading Trading Quant Finance Deep Learning Mathematics Linear Algebra Probability Optimisation Stochastic Calculus Time Series Both are parameter estimation under uncertainty. Both descend the same loss surface. A working tour of the shared mathematical principles — maximum likelihood, convex optimisation, covariance decomposition, regularisation, generalisation, stochastic calculus, Bayesian filtering, reinforcement learning — that make the two fields a single discipline with two vocabularies. — The vocabulary separating algo trading from machine learning is a historical accident. Strip it away and you find one subject: gradient descent on a regularised empirical risk, with the covariance matrix doing the heavy lifting and the Bellman equation governing sequential decisions. This essay writes the equations twice, in each field's dialect, and points at the object in the middle — ridge equals shrinkage, Kalman equals state-space RNN, Merton equals policy gradient, Feynman–Kac equals Anderson's reverse SDE. Where the analogy holds, where it breaks, and why the practitioner who masters the shared math has a larger edge than the one who masters the framework. https://hasanjaved.me/blog/lightgbm-still-the-best-model-for-trading/ https://hasanjaved.me/blog/lightgbm-still-the-best-model-for-trading/ Mon, 20 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) AI Algo Trading Trading Quant Finance Deep Learning Python Production In the age of transformers and frontier LLMs, the unfashionable truth is that most trading signals still live in gradient-boosted trees. A practitioner's defence of a six-year-old model that keeps winning. — Three weeks ago a friend killed a transformer after it lost money live against the LightGBM it was supposed to replace. Nobody was surprised. This is an essay about why, on small, noisy, non-stationary data, humble models keep beating fashionable ones — and the specific workflow that makes LightGBM the default first choice on a 2026 trading desk. https://hasanjaved.me/blog/why-django-is-still-the-best-backend-framework/ https://hasanjaved.me/blog/why-django-is-still-the-best-backend-framework/ Sun, 19 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Backend Django Backend Python AI Fintech Production Essay The prevailing wisdom is that Django is old. Old like a Corolla — unfashionable, unglamorous, and still the thing that gets you to work every morning. A contrarian essay on why the least exciting framework in Python is the one most likely to ship your AI product to production. — FastAPI won the async argument. Next.js won the fullstack argument. Rails won the ergonomics argument. Django just kept shipping. In the age of LLM apps that assemble a dozen vendor APIs into a useful product, the batteries-included framework is not the nostalgic choice — it's the pragmatic one. https://hasanjaved.me/blog/top-five-strategies-eloquent-math/ https://hasanjaved.me/blog/top-five-strategies-eloquent-math/ Sat, 18 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance Trading Algo Trading Quant Finance Stochastic Calculus Time Series Kelly Mathematics Econometrics Momentum, mean reversion, market making, volatility risk premium, and factor investing — each distilled to the equation that actually makes the trade, illustrated with the graph that shows why. — Trading strategies are not infinite. Five families account for almost every durable quant return stream of the last half-century. This essay takes each one, writes the math as the original paper wrote it, shows the graph that makes the math obvious, and — because a strategy without its failure modes is a prospectus — names the precise conditions under which each one stops working. https://hasanjaved.me/blog/modern-finance-dashboard-on-a-budget/ https://hasanjaved.me/blog/modern-finance-dashboard-on-a-budget/ Fri, 17 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance DevOps Infrastructure Cost Engineering DuckDB NautilusTrader Parquet Python Trading Fintech Eighteen months of open-source maturation collapsed the stack. The before-and-after, the new line items, a worked cost model, and the honest places the cheap stack still breaks. — A small quant shop asked me to price the same dashboard they paid $4,800 a month for in 2024. The 2026 estimate came in under $140. Most of the compression wasn't ingenuity — it was patient substitution of paid middleware with open-source equivalents that were good-enough then and are production-grade now. https://hasanjaved.me/blog/figma-adobe-stock-linkage-deep-dive/ https://hasanjaved.me/blog/figma-adobe-stock-linkage-deep-dive/ Fri, 17 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance Econometrics Statistics Trading Markets Pairs Trading Cointegration Time Series Granger Causality Three years after the $20B deal collapsed, the market has had long enough to tell us whether ADBE and FIG move as rivals, twins, or merely sector-mates. A correlation study, a cointegration test, and the statistical rigor the question deserves. — Adobe tried to buy Figma, then didn't. Figma IPO'd, then soared. Three years and a $1B breakup fee later, both tickers trade on the same screens — and retail investors keep asking whether they move together. This essay answers in the only way the question can be honestly answered: with data, formulas, and the humility to name what the numbers don't tell us. https://hasanjaved.me/blog/essential-mathematics-for-quants/ https://hasanjaved.me/blog/essential-mathematics-for-quants/ Fri, 17 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Mathematics Mathematics Quant Finance Probability Stochastic Calculus Linear Algebra Optimisation Information Theory Kelly A working quant's toolkit — probability, linear algebra, optimisation, stochastic calculus, time series, information theory, and numerics — and the algorithms each of them makes possible. — Strip away the Python, the Rust, and the execution venues, and the quant job is mathematics. Seven branches, properly rendered, and the algorithms each one quietly powers — from factor models to Black–Scholes to the Kelly criterion. https://hasanjaved.me/blog/introduction-to-algorithmic-trading-three-things/ https://hasanjaved.me/blog/introduction-to-algorithmic-trading-three-things/ Thu, 16 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance Algo Trading Trading Risk Management Kelly Backtesting Foundations The three lessons that separate strategies that survive from strategies that deliver a tuition bill — written for the trader who has just started and would like to skip the expensive parts. — Everyone starts algo trading thinking the hard part is the strategy. It isn't. The hard part is data quality, execution realism, and position sizing — and each of these will silently delete your account if you don't understand them. Here's what the textbooks don't lead with. https://hasanjaved.me/blog/best-python-backtesting-libraries-2026/ https://hasanjaved.me/blog/best-python-backtesting-libraries-2026/ Thu, 09 Apr 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance Backtesting Algo Trading Trading Quant Finance Python Rust NautilusTrader NautilusTrader, VectorBT, Backtrader, Zipline, QuantLib, LEAN — a field guide from someone who has shipped strategies on all of them. — There is no single best backtester; there is only the best backtester for the shape of your problem. Event-driven vs vectorized, latency-honest vs fast-to-iterate, crypto vs equities. A comparison written by an engineer who has run real money on each of these frameworks. https://hasanjaved.me/blog/nautilus-trader-production-field-report/ https://hasanjaved.me/blog/nautilus-trader-production-field-report/ Thu, 19 Mar 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance NautilusTrader Rust Event-Driven Execution Live Trading Trading Production DevOps What happens when you move an event-driven trading stack from Python to a Rust core — and the specific pain you save (and don't). — A year of running NautilusTrader against live FX and crypto venues. Where the Rust core earns its keep, where the Python API still leaks abstractions, and the three configuration mistakes that will eat your P&L if you don't catch them early. https://hasanjaved.me/blog/is-mathematics-invented-or-discovered/ https://hasanjaved.me/blog/is-mathematics-invented-or-discovered/ Fri, 27 Feb 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Mathematics Mathematics Philosophy Foundations Essay A 2,500-year argument about whether numbers live inside our skulls or somewhere beyond them. — Platonists say the primes were already there, waiting. Formalists say we stitched them into being out of rules and chalk dust. Neither camp has ever fully won — and that, perhaps, is the clue. https://hasanjaved.me/blog/the-saas-downfall/ https://hasanjaved.me/blog/the-saas-downfall/ Thu, 22 Jan 2026 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Finance SaaS Markets Strategy AI Production Essay SaaS isn't dying. Undifferentiated SaaS is. The survivors look like banks — regulated-vertical, services-as-a-software, or distribution-owner. Three archetypes, named examples, the math of each. — The 2018 SaaS playbook is dead. What replaces it is a sorting: every survivor falls into one of three archetypes — regulated-vertical (Veeva, Toast, Procore), services-as-a-software (Cresta, Harvey, Sierra), or distribution-owner (Shopify, HubSpot, Klaviyo). Each one looks more like a specialty bank than a startup. The undifferentiated horizontal middle is the casualty. https://hasanjaved.me/blog/how-neural-networks-use-elegant-mathematics/ https://hasanjaved.me/blog/how-neural-networks-use-elegant-mathematics/ Thu, 11 Dec 2025 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) AI AI Deep Learning Backpropagation Attention Calculus Mathematics Linear Algebra From the chain rule to attention — the quiet equations that make modern AI work. — Strip away the GPUs and the hype and what remains is surprisingly small: a dot product, a nonlinearity, a gradient, and the chain rule applied with absurd patience. Here's the math, rendered properly. https://hasanjaved.me/blog/risk-and-return/ https://hasanjaved.me/blog/risk-and-return/ Mon, 29 Jan 2024 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Fundamentals Essay Foundations Markets Risk Management Statistics Quant Finance Volatility, drawdown, illiquidity, ruin. Four different faces of risk, four different graveyards. The investor's job is not to maximise risk-adjusted return — it is to name which face is paying them. — Sharpe ratios collapse four very different things — volatility, drawdown, illiquidity, ruin — into one number. LTCM was killed by illiquidity, not volatility. Short-vol funds were killed by drawdown, not vol. FTX customers were killed by ruin, invisible until the day. Three graveyards, one lesson: name the face of risk that is paying you, or you are speculating. https://hasanjaved.me/blog/compounding-from-first-principles/ https://hasanjaved.me/blog/compounding-from-first-principles/ Mon, 22 Jan 2024 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Fundamentals Essay Foundations Mathematics Probability The one equation that runs a surprising amount of the world — demystified with clean math, worked examples, and the practical conclusions nobody tells you when you're twenty-two. — Einstein probably never called compound interest the eighth wonder of the world. The misattribution persists because the statement is nearly true. Compounding is the reason a small sum saved in your twenties beats a large sum saved in your forties, the reason credit-card debt is ruinous, and the reason nobody gets rich slowly enough to notice. Here's the math, cleanly, with the practical conclusions that follow. https://hasanjaved.me/blog/what-is-a-stock/ https://hasanjaved.me/blog/what-is-a-stock/ Mon, 15 Jan 2024 00:00:00 GMT hasanjaved065@gmail.com (Hasan Javed) Fundamentals Essay Foundations Markets Fintech Not a ticker. Not a number on a screen. A legal claim on the cash flows and governance of a real business — and understanding it that way changes every subsequent decision you make as an investor. — Most first-time investors think of a stock as a number that goes up and down. It is not. It is a share in a business: a slice of ownership that confers a claim on future profits, a vote on governance, and a position at the back of the line if things go wrong. Once you see it that way, the rest of investing makes more sense.