THEMIS
Financial Data Engineering & Point-in-Time Architecture
Diseño de Data Centers Financieros Multi-Asset para Research, Backtesting y Trading Algorítmico
$2,200
Mission and vision
Mission
Design Financial Data Truth Infrastructures that preserve meaning, traceability, temporality, provenance, and scientific reproducibility — answering what datum existed, what it meant, and when it could actually be known.
Vision
Practitioners who can design a multi-asset, provider-agnostic, Point-in-Time data architecture that preserves history, corrections, identity, temporal availability, provenance, and reproducible data evidence without contaminating research with look-ahead.
Description
THEMIS is Financial Data Engineering & Point-in-Time Architecture. It teaches how to design a Financial Data Center that preserves meaning, traceability, temporality, provenance, and scientifically defensible data evidence.
The object is not downloading prices or storing OHLCV. The object is a Financial Data Truth Infrastructure that can answer: what datum existed, what it meant, and when it could actually be known.
Who it is for
Algorithmic traders, quantitative researchers, trading-system developers, financial data engineers, quants who build backtests, researchers applying machine learning to markets, platform developers, quantitative instructors, and professionals working with futures, FX, equities, macro, and multi-asset research.
Prerequisites
Intermediate Python, basic DataFrames, fundamental market concepts, OHLCV and time series, elementary backtesting. Basic SQL is useful. Distributed systems, cloud, and microservices are not required.
Professional outcome
Design a multi-asset, provider-agnostic, Point-in-Time financial data architecture that preserves history, corrections, identity, temporal availability, provenance, and scientific reproducibility without contaminating research with look-ahead.
Distinctive technical processes
Raw → Canonical → Derived without silent vendor merges. Multiple financial clocks (`event_time`, `release_time`, `available_at`, `received_at`, `ingested_at`, `revision`). Availability claims (DETERMINED / IMPUTED / UNDETERMINABLE). Independent observation entities instead of a fat daily bar. Sovereign instrument identity and trading calendars. Futures contracts as canonical, continuous series as derived. Corporate-action events versus adjustment factors. Macro vintages. Quality that annotates and does not edit. Sealed datasets, manifests, hashes, M1 / M2 / M3, auditability, and re-executability.
Relation to quantitative systems
THEMIS is a sovereign Financial Data Infrastructure. It is a natural upstream for quantitative systems, including ZEUS ENGINE, because it produces the data-truth capabilities those systems need. ZEUS ENGINE does not depend on THEMIS internals. The correct dependency is data contracts: Financial Data Infrastructure (THEMIS or another compatible system) → Data Contract Boundary → ZEUS ENGINE.
THEMIS may also feed other lines. The curricular relation is a graph, not a single public chain: ZEUS ENGINE, Regime Intelligence, Pairs Trading, Multicointegration, Data Mining, Machine Learning, and other quantitative systems.
Module 12 is not CHRONOS
THEMIS M12 is Reproducible Data Infrastructure: sealed datasets, manifests, hashes, PIT data delivery, versioned policies, M1 / M2 / M3, auditability, re-executability.
THEMIS does not teach Temporal Air Gap, SV3 as research-governance, research cutoff, experimental worlds, research reopening, decision admissibility, or CHRONOS as a phase. Those belong to ZEUS / CHRONOS (Temporal Research Governance).
THEMIS delivers temporally defensible data evidence. CHRONOS governs permitted knowledge inside the experiment.
Professional Capstone
THEMIS — Professional Capstone. Implementation identity: Financial Data Center Reference Implementation. Controlled scope (one instrument, one futures or equity lifecycle, one revisable macro series, one provider abstraction, Raw, Canonical, PIT query, quality, sealed snapshot) does not reduce architectural rigor. It is not a toy, a mini-lab, or a demo architecture.
Canonical syllabus: `courses/themis/SYLLABUS.md`. This page is a mount.
What you'll be able to do
- Design a multi-asset, provider-agnostic, Point-in-Time financial data architecture
- Separate Raw, Canonical, and Derived without silently merging vendors or destroying history
- Model financial clocks and PIT data delivery — a date filter is not a PIT architecture
- Preserve sovereign identity, calendars, futures contracts, corporate actions, and macro vintages
- Deliver sealed, hashed, re-executable data evidence (M12) — not experimental research governance
Course content
Phase I — FOUNDATIONS
Module 1 — Financial Data Center Architecture
Module 2 — Provider & Provenance Engineering
Phase II — PIT
Module 1 — Financial Time Semantics
Module 2 — Point-in-Time Architecture
Phase III — SEMANTICS
Module 1 — Financial Observation Model
Module 2 — Market Micro-Semantics
Phase IV — REFERENCE
Module 1 — Instrument Identity & Symbology
Module 2 — Calendars & Sessions
Phase V — DOMAINS
Module 1 — Futures Engineering
Module 2 — Corporate Actions & Macroeconomic Vintages
Phase VI — GOVERNANCE
Module 1 — Quality, Corrections & Lineage
Module 2 — Reproducible Data Infrastructure
THEMIS — Professional Capstone
Course content is Phase → Module → Chapter using THEMIS's six phases (FOUNDATIONS → PIT → SEMANTICS → REFERENCE → DOMAINS → GOVERNANCE). Module numbers restart inside each phase. Chapter IDs are P.M.C.- like ZEUS. THEMIS — Professional Capstone sits outside the 12 / 36 count. Lecture bodies are not authored yet.
Instructor
Elmer Niño
Founder, VTAlgo Group · Systems Engineer · MSc in Computer Science
Algorithmic trader since 2009, running trading as an enterprise. Methodology built on walk-forward analysis, genetic optimization, PCA against over-optimization, the Triple Barrier Method, and 9–12 month live validation before capital.
THEMIS
Financial Data Engineering & Point-in-Time Architecture
Diseño de Data Centers Financieros Multi-Asset para Research, Backtesting y Trading Algorítmico
$2,200
Misión y visión
Misión
Design Financial Data Truth Infrastructures that preserve meaning, traceability, temporality, provenance, and scientific reproducibility — answering what datum existed, what it meant, and when it could actually be known.
Visión
Practitioners who can design a multi-asset, provider-agnostic, Point-in-Time data architecture that preserves history, corrections, identity, temporal availability, provenance, and reproducible data evidence without contaminating research with look-ahead.
Description
THEMIS is Financial Data Engineering & Point-in-Time Architecture. It teaches how to design a Financial Data Center that preserves meaning, traceability, temporality, provenance, and scientifically defensible data evidence.
The object is not downloading prices or storing OHLCV. The object is a Financial Data Truth Infrastructure that can answer: what datum existed, what it meant, and when it could actually be known.
Who it is for
Algorithmic traders, quantitative researchers, trading-system developers, financial data engineers, quants who build backtests, researchers applying machine learning to markets, platform developers, quantitative instructors, and professionals working with futures, FX, equities, macro, and multi-asset research.
Prerequisites
Intermediate Python, basic DataFrames, fundamental market concepts, OHLCV and time series, elementary backtesting. Basic SQL is useful. Distributed systems, cloud, and microservices are not required.
Professional outcome
Design a multi-asset, provider-agnostic, Point-in-Time financial data architecture that preserves history, corrections, identity, temporal availability, provenance, and scientific reproducibility without contaminating research with look-ahead.
Distinctive technical processes
Raw → Canonical → Derived without silent vendor merges. Multiple financial clocks (`event_time`, `release_time`, `available_at`, `received_at`, `ingested_at`, `revision`). Availability claims (DETERMINED / IMPUTED / UNDETERMINABLE). Independent observation entities instead of a fat daily bar. Sovereign instrument identity and trading calendars. Futures contracts as canonical, continuous series as derived. Corporate-action events versus adjustment factors. Macro vintages. Quality that annotates and does not edit. Sealed datasets, manifests, hashes, M1 / M2 / M3, auditability, and re-executability.
Relation to quantitative systems
THEMIS is a sovereign Financial Data Infrastructure. It is a natural upstream for quantitative systems, including ZEUS ENGINE, because it produces the data-truth capabilities those systems need. ZEUS ENGINE does not depend on THEMIS internals. The correct dependency is data contracts: Financial Data Infrastructure (THEMIS or another compatible system) → Data Contract Boundary → ZEUS ENGINE.
THEMIS may also feed other lines. The curricular relation is a graph, not a single public chain: ZEUS ENGINE, Regime Intelligence, Pairs Trading, Multicointegration, Data Mining, Machine Learning, and other quantitative systems.
Module 12 is not CHRONOS
THEMIS M12 is Reproducible Data Infrastructure: sealed datasets, manifests, hashes, PIT data delivery, versioned policies, M1 / M2 / M3, auditability, re-executability.
THEMIS does not teach Temporal Air Gap, SV3 as research-governance, research cutoff, experimental worlds, research reopening, decision admissibility, or CHRONOS as a phase. Those belong to ZEUS / CHRONOS (Temporal Research Governance).
THEMIS delivers temporally defensible data evidence. CHRONOS governs permitted knowledge inside the experiment.
Professional Capstone
THEMIS — Professional Capstone. Implementation identity: Financial Data Center Reference Implementation. Controlled scope (one instrument, one futures or equity lifecycle, one revisable macro series, one provider abstraction, Raw, Canonical, PIT query, quality, sealed snapshot) does not reduce architectural rigor. It is not a toy, a mini-lab, or a demo architecture.
Canonical syllabus: `courses/themis/SYLLABUS.md`. This page is a mount.
What you'll be able to do
- Design a multi-asset, provider-agnostic, Point-in-Time financial data architecture
- Separate Raw, Canonical, and Derived without silently merging vendors or destroying history
- Model financial clocks and PIT data delivery — a date filter is not a PIT architecture
- Preserve sovereign identity, calendars, futures contracts, corporate actions, and macro vintages
- Deliver sealed, hashed, re-executable data evidence (M12) — not experimental research governance
Course content
Phase I — FOUNDATIONS
Module 1 — Financial Data Center Architecture
Module 2 — Provider & Provenance Engineering
Phase II — PIT
Module 1 — Financial Time Semantics
Module 2 — Point-in-Time Architecture
Phase III — SEMANTICS
Module 1 — Financial Observation Model
Module 2 — Market Micro-Semantics
Phase IV — REFERENCE
Module 1 — Instrument Identity & Symbology
Module 2 — Calendars & Sessions
Phase V — DOMAINS
Module 1 — Futures Engineering
Module 2 — Corporate Actions & Macroeconomic Vintages
Phase VI — GOVERNANCE
Module 1 — Quality, Corrections & Lineage
Module 2 — Reproducible Data Infrastructure
THEMIS — Professional Capstone
Course content is Phase → Module → Chapter using THEMIS's six phases (FOUNDATIONS → PIT → SEMANTICS → REFERENCE → DOMAINS → GOVERNANCE). Module numbers restart inside each phase. Chapter IDs are P.M.C.- like ZEUS. THEMIS — Professional Capstone sits outside the 12 / 36 count. Lecture bodies are not authored yet.
Instructor
Elmer Niño
Founder, VTAlgo Group · Systems Engineer · MSc in Computer Science
Algorithmic trader since 2009, running trading as an enterprise. Methodology built on walk-forward analysis, genetic optimization, PCA against over-optimization, the Triple Barrier Method, and 9–12 month live validation before capital.