Artificial intelligence is moving faster than conventional courses can be rewritten or busy financial professionals can independently explore. This living educational platform continuously investigates what is changing, tests its relevance to financial problems, and translates the results into rigorous, accessible education for institutions and individuals.
AI & Algorithmic Finance — A Living Curriculum
Choose your audienceGraduate Students · Financial Practitioners · Boards & LeadershipSelect a domain to explore its audience-specific roadmap and original educational resources.
Traditional courses are episodic. A frontier-facing curriculum must be living: stable where principles endure, revised where architectures change, and exploratory where important new capabilities appear. Classroom teaching, including graduate-level instruction, provides an ongoing opportunity to test explanations and laboratories; each new iteration informs the next.
Preserve economic reasoning, statistical discipline, market mechanisms, robust evaluation, controls and professional accountability. Revise when evidence—not fashion—requires it.
Revisit courses and experiments as machine learning gives way to conversational models, tool-using agents, coordinated systems and more adaptive architectures.
Investigate consequential developments early, distinguish promise from demonstrated capability, and translate experiments into financial use cases with explicit uncertainty.
The same technological advance matters differently to a board director, a professional managing a financial workflow, and a graduate student preparing to enter the field. The platform adapts depth, cases and expected outcomes—not the underlying intellectual standards.
Strategic implications, governance, institutional readiness, fiduciary oversight, model risk, delegated authority and responsible adoption.
Trading, investment banking, corporate finance, asset management, insurance, risk and regulation—translated into relevant workflows, practical laboratories and implementation choices.
Progressive foundations, conceptual bridges, guided experiments, reproducible notebooks and the judgment required for professional practice.
The author identifies questions at the intersection of emerging AI and financial practice, uses AI to accelerate exploration and prototyping, tests mechanisms through executable laboratories, and curates the lessons that professionals need. The objective is not to outsource judgment to AI, but to use AI to investigate better questions and expose the limits of proposed answers.
Scan new models, architectures and practices; ask which advances materially change a financial task, decision or institutional control.
Connect monographs, lecture decks, podcasts and runnable Colab notebooks. Make assumptions, intermediate steps, failure modes and governance visible.
Move toward periodic frontier editions that explicitly state what remains valid, what requires revision, and what deserves fresh investigation. This is the evolving delivery model, not a claim of an existing paid service.
Quantum-inspired finance, engineered AI systems and structural reasoning enlarge the toolkit for explaining emerging financial applications. The following existing research cards remain part of the complete catalogue.
Classical methods inspired by quantum computing, geometry, and structured optimization — applied to portfolios, regimes, risk, and decision systems with verifiable artifacts.
Engineering-first AI for finance: reproducible notebooks, explicit assumptions, structured outputs, and governance gates for institutional-quality work.
Reasoning treated as a designed system: artifacts, roles, verification discipline, and explicit uncertainty — scaling capability without scaling risk.
Start with a domain, select your audience, and follow a numbered learning roadmap. Each chapter connects to the original courses, runnable laboratories and research resources. The reference and research libraries below offer complementary depth, evidence and context.
The existing teaching catalogue is organized into two complementary domains: AI Systems (governance-first applied AI) and Trading Systems (systematic trading, market mechanisms, and governed laboratories). Both domains share the same operating discipline: runnable work, explicit assumptions, and auditable artifacts.
Course-books and notebook stacks for high-accountability professional AI use: scope control, verification posture, artifact logging, risk controls, and defensible implementation patterns across regulated workflows.
All original resources remain available. Select an audience to see a recommended conceptual sequence; no cards or links are removed.
Foundations, guided experiments and critical judgment.
From intelligence and institutional knowledge to experience-based judgment. A research monograph and pedagogical resources exploring governed AI memory, validated lessons from financial decisions, selective forgetting, and the feedback architectures that turn operational outcomes into accountable institutional learning.
A compact, book-supported learning ecosystem for financial professionals who want to understand modern AI without becoming computer scientists. Built around 20 essential chapters, 20 Colab notebooks, 20 audio presentations, and 20 specialized decks, it explains neural networks, deep learning, transformers, embeddings, reasoning models, agents, and generative AI with practical clarity and governance-first discipline.
A foundational programme in AI Systems: a monumental self-teaching course designed specifically for Master of Finance students and financial professionals. Built as a four-layer learning architecture — books, decks, videos and podcasts, and executable Colab notebooks — this program treats artificial intelligence not as a mere software tool, but as industrial production infrastructure requiring disciplined capital allocation, risk management, and rigorous institutional governance.
The canonical entry point to the full Governance-First training ladder: Machine Learning foundations → governed generative AI operations → fine-tuning governance → frontier governance. Built for auditability, reproducibility, and high-accountability professional practice.
The flagship curriculum connecting AI-driven finance with quantum-inspired thinking — built as an executable course (books + releases + Colab notebooks) with governance-first discipline.
A mechanics-first laboratory for learning the operational foundations of machine learning and artificial intelligence through executable notebooks and chapter-linked materials. Built as a structured bridge from conceptual understanding to runnable implementation across core architectures, with emphasis on clarity, reproducibility, and educational depth.
The foundational volume: governance discipline before GenAI—scope, verification posture, audit artifacts, review gates, and human accountability for ML workflows.
A governance-first course-book and governed notebook stack for prompt design in professional workflows. Treats prompting as a specification discipline: explicit scope, role definition, boundary control, schema enforcement, staged prompting, and auditable artifacts — with outputs designed for review, traceability, and accountable use.
A governance-first, book-supported course with an executable Colab notebook stack — designed for defensible, reviewable AI use in banking workflows and institutional implementation.
Governance-first courseware for consulting and corporate strategy workflows — built for defensible practice, reusable assets, and auditable artifacts.
Governance-first courseware for U.S. wealth management and financial advice — designed for supervised, reviewable outputs and evidence discipline.
Governance-first courseware for U.S. audit and accounting practice — capability ↑ ⇒ risk ↑ ⇒ controls ↑, with auditable artifacts and lab workflows.
Governance-first practical guide for legal workflows — explicit uncertainty, verification discipline, and zero tolerance for invented authority.
A governance-first course-book: disciplined fine-tuning for regulated workflows. Treats fine-tuning as institutional governance — task scope, forbidden behaviors, evaluation, approval gates, audit artifacts, monitoring, rollback, and human accountability (with runnable Colab notebooks).
Two-volume governance-first exploration of frontier AI implementation. Covers emerging architectures, agentic systems, reasoning models, multimodal risk, and institutional control patterns — grounded in real deployment constraints and paired with executable Colab notebooks.
A governance-first, book-supported course with executable Colab notebooks for learning Bayesian thinking in AI: probabilistic inference, prior-to-posterior updating, uncertainty-aware prediction, Bayesian neural networks, and disciplined decision-making under incomplete information. Built for runnable understanding, explicit assumptions, and auditable educational artifacts.
A governance-first, book-supported course with executable Colab notebooks for understanding diffusion models through probabilistic noising and denoising, iterative generation, reverse-process learning, and controlled reconstruction. Built for runnable learning, explicit assumptions, auditable artifacts, and disciplined interpretation in professional and educational settings.
A governance-first, Colab-first handbook that teaches LangGraph-only agentic architectures for finance: explicit state, routing, bounded loops, termination rules, and reviewable artifacts. Built to produce committee-ready workflows that explain what happened, why it happened, and what needs human sign-off.
A governance-first laboratory for building “surrogate” AI roles that stand in for specialist reviewers (risk, compliance, PM, analyst) to stress-test drafts and decisions. Focused on structured critique, disagreement logging, escalation paths, and auditable artifacts documenting assumptions, evidence, and red flags.
A governance-first multimodal analysis stack combining text with images (and optionally other modalities) for finance and professional workflows. Emphasis is contract-like pipelines: explicit inputs, constraints, routing, and termination, with artifacts that audit provenance and interpretability.
A governance-first approach to long-context memory: storing, retrieving, and using prior information safely in professional workflows. Emphasizes bounded recall, citation-aware retrieval, and strict separation of facts vs assumptions to prevent “memory hallucinations,” with full memory-use logs.
A governance-first course-book and governed notebook stack for reasoning architectures (chains, trees, loops, committees). Treats reasoning as controlled inference: explicit state, routing, constraints, termination, stage gates, and auditable artifacts — with outputs explicitly marked Not verified until independently reviewed.
A five-domain AI for Science collection applying governed discovery loops to high-accountability financial and professional workflows: algorithmic trading, tax planning, civil litigation strategy, household financial advice, and investment banking / M&A. Each domain follows the same architecture: the model proposes, deterministic engines compute, stress tests challenge, professional review agents scrutinize, and evaluation agents decide whether to continue, stop, or revise.
A pedagogical journey from understanding to governed creation.
An Integrated Autonomous Intelligence teaching and research collection
The decisive frontier in artificial intelligence is no longer the isolated quality of a model’s answer. It is the design of systems that can pursue objectives across time: systems that remember, retrieve, reason, use tools, coordinate workflows, update persistent state, and participate in consequential institutional work.
From protected knowledge vaults to governed cognitive evolution.
An end-to-end framework for connecting protected organizational knowledge to AI while preserving control over proprietary information. Governed retrieval, provenance, deterministic controls, policy rules, and human approval determine what evidence models may use and what candidate knowledge may enter institutional memory.
From foundational cyber risk to governed defense against agentic threats.
A book-supported course and governed Safe Lab series for financial practitioners, students, executives, and boards. Twenty synthetic Colab laboratories progress from basic financial-integrity controls to prompt injection, adaptive red–blue interaction, poisoned memory, false consensus, deterministic containment, and causal recovery.
From spontaneous agent collaboration to strategic diagnosis and governed intervention.
A three-paper research journey and ten-laboratory computational collection exploring how autonomous AI agents interact, coordinate, and form strategic structures. Cooperative and non-cooperative game theory, causal diagnosis, inverse-game inference, intervention, and provenance are combined to distinguish genuine collaboration from correlation and translate emergent multi-agent behavior into defensible institutional action.
From self-improving procedures to adaptive organizations and artificial societies.
A collection of monographs and Colab laboratories exploring how autonomy extends into procedures, teams, relationships, rules, and organizational boundaries. Self-improving optimizers, recursive organizational adaptation, swarm discovery, constitutional swarms, and cross-society collaboration form a methodological ladder grounded in bounded experiments, governance, and auditable evidence.
Applying AI swarms to financial research, discovery, decision systems, and institutional applications.
A five-episode research and teaching series built around swarms of 50 heterogeneous AI agents tackling progressively more complex financial problems: algorithmic investing, American option research, global insurance architecture, cross-science market regime discovery, and the autonomous financial institution. The project focuses on substantive heterogeneity, preserved disagreement, external truth surfaces, deterministic constraints, provenance, and governed organizational adaptation.
From capable models to governed financial institutions.
A pedagogical research paper and six Colab notebooks exploring the layers of AI harnessing: reliability, authority, deliberation, action, adaptation, and evidence. Accounting, banking regulatory approvals, central-bank decisions, institutional trading, bank liquidity crises, and syndicated-credit committees provide the financial settings. The series shows how controlled data, deterministic tools, validation, decision rights, state-dependent permissions, and provenance turn model intelligence into accountable institutional processes.
A compact executive learning architecture for CEOs, senior financial practitioners, and advanced graduate students who need to understand how modern artificial intelligence is evolving from statistical and machine-learning systems into generative AI, agents, constellations, and autonomous architectures. The course combines monographs, decks, infographics, and executable notebooks to explain how different forms of intelligence work together across perception, judgment, tools, execution, feedback, and governance.
A five-laboratory learning architecture for organizing knowledge and assembling decision-relevant context. Obsidian-style vaults, typed knowledge graphs, source provenance, temporal validity, and bounded AI judgment support applications in regulatory intelligence, market contagion, M&A discovery, institutional memory, and legal research.
Systematic trading treated as an engineered control system: environment → signal → policy → execution → feedback. These course-books are runnable, mechanism-first, and explicit about costs, feasibility, regimes, and stress behavior.
All original resources remain available. Select an audience to see a recommended conceptual sequence; no cards or links are removed.
Foundations, guided experiments and critical judgment.
A self-teaching course in algorithmic trading and algorithmic systems built as a multilevel formation framework: book → decks → podcasts → videos → executable Colab notebooks. Emphasizes systems engineering, governed strategy design, execution realism, market mechanisms, and progressive professional formation.
A practical supplement to the Self-Teaching Algorithmic Trading platform. Rather than a single laboratory, this repository teaches students how to build self-contained algorithmic trading laboratories: complete frameworks for exploring, testing, governing, and producing stable, defensible algo platforms across equities, FX, commodities, credit, and volatility markets. The emphasis is on system construction, agentic architecture, auditability, stress testing, calibration, and professional research discipline.
A unified formation framework: discipline → exploration → implementation. Emphasizes constraint surfaces, reproducibility, execution realism, and human accountability.
An executable notebook stack focused on professional trading engineering: honest backtesting, ML integration, portfolio construction, execution realism, and governance artifacts.
A synthetic-first strategy laboratory: canonical archetypes implemented under governance, with auditable artifacts, stress tests, feasibility cliffs, and explicit “Not verified” posture.
A mechanism-first primer on how markets clear: microstructure, liquidity, execution friction, and regime behavior — built as runnable, governed learning assets for trading practitioners.
A quantum-inspired finance learning platform focused on the essential computational models for trading and investing: quantum states, portfolio optimization, Hamiltonian representations, governance-first Shor workflows, quantum neural networks, hybrid quantum-classical modeling, and agentic financial discovery. Built as an educational bridge between quantum algorithms, systematic investing, and governed financial research.
A complete research and teaching journey in algorithmic trading in the age of artificial intelligence: book → executable notebooks → QuantConnect implementation → ChatGPT-based autonomous research assistant. Connects machine learning, strategy design, portfolio construction, execution realism, independent risk, and auditability within a governed architecture that preserves research discipline and human accountability.
The long-form layer: books and deep research write-ups that develop ideas beyond a single paper. The focus remains the same: AI-driven finance, quantum-inspired thinking, and governance-first engineering.
Long-format books and monographs across AI, finance, governance, and quantum-inspired approaches — designed as durable references.
Start with a monograph, then move into runnable course notebooks when you want execution artifacts. Use papers and pedagogical monographs in the Research Library when you need citable results or a structured conceptual journey.
All work is built in public with open-source tooling. If you want to collaborate, start from the GitHub profile and repositories.
Academic papers provide focused results, working drafts, and applied studies. Pedagogical monographs develop the conceptual architecture in a more structured and accessible form, connecting theory, governance, and application. Both are designed to be inspectable, citable, and closely connected to the long-form books and executable courses.
Working papers and applied research on engineered intelligence, structural reasoning, and financial modeling.
A curated collection of pedagogical monographs focused primarily on autonomous systems and institutional knowledge. The collection develops the architectures, concepts, governance principles, and applications required to understand how intelligent systems can evolve into accountable institutional capabilities.
Academic papers provide modular results, pedagogical monographs build structured conceptual journeys, Long-form books develop the deep narrative and theory, and Executable courses provide executable implementations and artifacts.
When a paper introduces a method, the goal is to make it runnable (or align it) with notebook workflows: deterministic runs, logged assumptions, and explicit artifacts.
Interested in supporting or collaborating on a living educational platform for AI and algorithmic finance, including teaching pathways, frontier research, executable laboratories and responsible institutional adoption? I welcome partnerships across academia, industry, and innovation ecosystems.
Email: areynoso@yahoo.com
GitHub: github.com/alexdibol
Educational purpose only: All content across this site and the linked repositories (including books, notebooks, code, examples, templates, and explanatory text) is provided for educational and research purposes only. Nothing here constitutes investment advice, trading advice, financial advice, legal advice, tax advice, accounting/audit advice, compliance advice, or any other professional service. You should not rely on any material here to make decisions that require professional judgment. If you need advice, consult appropriately qualified professionals and follow applicable laws, regulations, and firm policies.
No warranties; use at your own risk: The materials are provided on an “AS IS” and “AS AVAILABLE” basis, without warranties of any kind (express or implied), including but not limited to warranties of accuracy, completeness, merchantability, fitness for a particular purpose, non-infringement, or availability. Methods, examples, and code may contain errors, may become outdated, and may not be suitable for your specific facts, jurisdiction, risk tolerance, constraints, or use case.
Limitation of liability: To the maximum extent permitted by law, Alejandro Reynoso disclaims all liability for any loss or damage of any kind (direct, indirect, incidental, consequential, special, exemplary, punitive, or otherwise) arising out of or related to the use of, inability to use, or reliance on these materials, including but not limited to trading losses, lost profits, business interruption, reputational harm, compliance failures, model or data errors, security incidents, or third-party claims.
Model / data / results are not verified: Unless explicitly stated otherwise in a particular artifact, treat all outputs, claims, calculations, citations, and conclusions as not verified. Users are responsible for independent verification, testing, and review before any real-world application. Past performance examples (if any) are illustrative and do not predict future results.
Security, confidentiality, and minimum-necessary data: Do not paste confidential, proprietary, regulated, or personally identifying information into external tools, prompts, or notebooks. Use anonymization/redaction and minimum-necessary inputs by default. You are responsible for complying with privacy, confidentiality, data handling, recordkeeping, and information security obligations.
Third-party tools and dependencies: Some notebooks or code may rely on third-party libraries, services, APIs, datasets, or model providers with their own licenses and terms. You are responsible for reviewing and complying with those terms and ensuring you have the right to use any referenced resources.
Generative AI assistance disclosure: Some portions of the content and/or code in these repositories may have been created with the assistance of generative AI tools. However, all overall structure, direction, organization, editorial supervision, governance framing, and final responsibility for the materials are human-led and remain with Alejandro Reynoso. Users should treat AI-assisted text or code as potentially fallible and must apply independent review and testing.
Copyright © Alejandro Reynoso. Repository-specific licenses and notices govern the linked materials.