# 3.14 Financial Contents

> We design and build systems that turn data and information into usable knowledge, content, predictions, decisions and actions, combining machine learning, generative models, knowledge engineering, agents, software and automation.

**We design and build systems that turn data and information into usable knowledge, content, predictions, decisions and actions.**

This index covers the four sections of the site. The full text of every page listed here is at https://www.3.14financialcontents.com/llms-full-en.txt.

Every link points to the Markdown version of a page. The equivalent HTML page is the same URL without the `index.md` suffix.

## Solutions

- [Solutions](https://www.3.14financialcontents.com/en/solutions/index.md): Intelligence to understand information better and to make the work advance.
- [Agents and orchestration](https://www.3.14financialcontents.com/en/solutions/agents-and-orchestration/index.md): Flexible reasoning inside operations with state, contracts and accountability.
- [Informational intelligence](https://www.3.14financialcontents.com/en/solutions/informational-intelligence/index.md): Turns scattered sources, data, language and logic into knowledge that can be understood, reused and put in motion.
- [Knowledge, research and evidence](https://www.3.14financialcontents.com/en/solutions/knowledge-research-and-evidence/index.md): Establishing what is known, what sources support it, which versions are in force and what remains open.
- [Generative and multimodal reporting](https://www.3.14financialcontents.com/en/solutions/informational-reporting-in-text-voice-and-video/index.md): From data and information to reporting in text, voice and video.
- [Decision intelligence](https://www.3.14financialcontents.com/en/solutions/decision-intelligence/index.md): Classifying, prioritising, anticipating, comparing scenarios and optimising decisions within objectives and constraints.
- [Language, translation and localisation](https://www.3.14financialcontents.com/en/solutions/language-translation-and-localisation/index.md): Systems that analyse, generate, translate and localise content while maintaining meaning, terminology, register and intent.
- [Operational intelligence](https://www.3.14financialcontents.com/en/solutions/operational-intelligence/index.md): Connects information, decisions, people, applications, agents and machines through to a verifiable result.
- [Operational intelligence in ERPs and processes](https://www.3.14financialcontents.com/en/solutions/operational-intelligence-in-erps-and-processes/index.md): Holding together work that crosses applications, rules, authorisations, waits and exceptions.
- [Knowledge intelligence for complex projects](https://www.3.14financialcontents.com/en/solutions/intelligence-for-complex-projects/index.md): Relating questions, evidence, decisions, dependencies and work to preserve the continuity of long projects.
- [Business logic and operational knowledge](https://www.3.14financialcontents.com/en/solutions/business-logic-and-operational-knowledge/index.md): Systems that reconstruct and make examinable the logic distributed across code, configurations, rules, documentation, data and operational practice.
- [Orchestration of agents and durable processes](https://www.3.14financialcontents.com/en/solutions/orchestration-of-agents-and-durable-processes/index.md): Systems that coordinate agents, software, tools and people through processes that must preserve state, wait, recover and continue to a verifiable result.

## Capabilities

- [Capabilities](https://www.3.14financialcontents.com/en/capabilities/index.md): The technical capabilities of our team.
- [Machine learning, optimisation and decision models](https://www.3.14financialcontents.com/en/capabilities/machine-learning-and-decision-models/index.md): We design models for classification, ranking, forecasting, anomaly detection, vision, time series, optimisation and decision-making over structured and unstructured data. We select models and methods according to the real problem, the…
- [AI architecture and advanced systems](https://www.3.14financialcontents.com/en/capabilities/technology/index.md): We design architectures for models, agents, inference, distributed systems, cloud and edge, runtimes and components of our own. We evaluate emerging technologies, build prototypes when necessary and take the ones that add value into…
- [Software and enterprise integration](https://www.3.14financialcontents.com/en/capabilities/software-and-enterprise-integration/index.md): We build applications, APIs, services, integrations, event-driven architectures and backend components to connect models, agents and knowledge systems with ERPs, databases, applications and existing processes. We also design the identity…
- [Information, semantics and knowledge](https://www.3.14financialcontents.com/en/capabilities/information-and-knowledge/index.md): We work with ingestion, structured extraction, entities, relationships, taxonomies, graphs, semantics, hybrid search, RAG, embeddings, reranking and memory. We design representations that make it possible to find, relate and reuse…
- [Linguistic engineering and knowledge-based generation](https://www.3.14financialcontents.com/en/capabilities/language-localisation-nlg/index.md): Analysing, representing, transforming and generating language under explicit constraints of meaning, terminology, context and audience.
- [Evaluation, observability and control](https://www.3.14financialcontents.com/en/capabilities/evaluation-observability-and-control/index.md): We design evals, adversarial testing, benchmarks, simulations, validators, traces, metrics and observability mechanisms for models, agents, knowledge systems, workflows, software and physical systems. We combine automatic evaluation…
- [Physical Intelligence](https://www.3.14financialcontents.com/en/capabilities/physical-intelligence/index.md): We work with computer vision, sensing, signal fusion, state estimation, simulation, navigation, motion planning, control, manipulation and edge computing. We explore how to integrate these capabilities with models, software and…

## Culture 3.14

- [Culture 3.14](https://www.3.14financialcontents.com/en/culture-314/index.md): Ask better. Build in order to learn. Unlearn when reality changes.
- [Building together](https://www.3.14financialcontents.com/en/culture-314/how-we-collaborate/index.md): We work with those who know, carry out, maintain, decide and receive the result. We frame the problem together, represent the domain, choose a first complete path and learn with artefacts that can be examined.
- [What we ask ourselves](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/index.md): Open questions that cut across information, semantics, language, decisions, operations, agents, projects, Physical Intelligence and method. Each one states why it matters, what we know and what remains open.
- [What we read](https://www.3.14financialcontents.com/en/culture-314/what-we-read/index.md): We collect books, papers, articles, projects and technical references that have changed a question, a design decision or the way we interpret a problem. Each reading explains what it moved in the work.
- [Laboratory](https://www.3.14financialcontents.com/en/culture-314/lab/index.md): Explorations, experiments, simulations, demonstrators, technical reviews of emerging capabilities and Physical Intelligence. Every piece of work declares its state of maturity.
- [Trajectory](https://www.3.14financialcontents.com/en/culture-314/trajectory/index.md): 3.14 began by working with information and went on to take in retrieval, language, statistics, machine learning, generation, automation, agents and integration. In the Laboratory it extends towards Physical Intelligence.
- [Physical operations](https://www.3.14financialcontents.com/en/culture-314/lab/physical-operations/index.md): The branch of the Laboratory where the system perceives and acts through sensors, controllers and physical mechanisms: vision, state estimation, navigation, planning, manipulation and edge.
- [Demonstrators](https://www.3.14financialcontents.com/en/culture-314/lab/demonstrators/index.md): Small systems, with public or synthetic data, that show a capability together with the conditions in which it works and the limits it still has.
- [Explore 3.14 and Ask 3.14](https://www.3.14financialcontents.com/en/culture-314/lab/explore-and-ask-314/index.md): Two ways of moving through this site: a discovery layer built on relations, and a bot limited to the public content that shows the pages it has used.

## Contact

- [Contact](https://www.3.14financialcontents.com/en/contact/index.md): Correo info@3.14financialcontents.com · teléfono +34911012001 · 3.14, Av. Esteiro, 145, 15403 Ferrol - A Coruña - Spain

## Optional

- [Evidence and structure come before composition](https://www.3.14financialcontents.com/en/culture-314/notes/content-should-not-start-with-the-writing/index.md): When the writing comes first, the evidence ends up fitting the text; when it comes last, the text fits the evidence.
- [Finding a source does not prove a conclusion](https://www.3.14financialcontents.com/en/culture-314/notes/finding-a-source-does-not-prove-a-conclusion/index.md): Retrieval measures proximity between texts; a conclusion requires a relation between what the source says and what is being claimed.
- [How can uncertainty be communicated without turning it into a misleading number?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-can-uncertainty-be-communicated-without-turning-it-into-a-misleading-number/index.md): 87 % confidence reads as a promise, even when the model is only saying that it has seen similar cases.
- [How do voice and video change what informational quality means?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/why-voice-and-video-change-the-quality-model/index.md): A text is reread; a narration is heard once and does not let the person receiving it go back to check.
- [How do you design an exception that no rule had anticipated?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-do-you-design-an-exception-no-rule-anticipated/index.md): The unforeseen cannot be enumerated, but you can decide in advance what happens to it.
- [How do you evaluate a shift in meaning that reads perfectly natural?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-do-you-evaluate-a-shift-in-meaning-that-reads-perfectly-natural/index.md): Automatic metrics reward fluency, and the most expensive error in a translation is precisely a fluent one.
- [How do you represent a contradiction without hiding it inside an answer?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-do-you-represent-a-contradiction-without-hiding-it-inside-an-answer/index.md): When two reliable sources say different things, summarising them into a single sentence destroys the most valuable information in the set.
- [How does a system know it has found enough evidence?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-does-a-system-know-it-has-found-enough-evidence/index.md): A system can retrieve ten relevant passages and still have nothing to support the claim it is about to write.
- [How should a system act when perception is not enough?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-should-a-system-act-when-perception-is-not-enough/index.md): Stopping has a cost, carrying on with a poor estimate has another, and the choice has to be made before it happens.
- [How to represent what a project does not yet know](https://www.3.14financialcontents.com/en/culture-314/notes/how-to-represent-what-a-project-does-not-yet-know/index.md): A task list records what has been agreed and loses whatever is still open, which is usually what decides the project.
- [Localising is adapting with invariants of meaning](https://www.3.14financialcontents.com/en/culture-314/notes/localising-is-not-substituting/index.md): A localisation decides what has to stay identical and lets everything else move; a substitution changes words and hopes the meaning survives.
- [Out of the Crisis · W. Edwards Deming](https://www.3.14financialcontents.com/en/culture-314/what-we-read/out-of-the-crisis-w-edwards-deming/index.md): Reacting to the normal variation of a process as though it were a signal makes the process worse; most errors come from the system and not from people.
- [Plans and Situated Actions · Lucy Suchman](https://www.3.14financialcontents.com/en/culture-314/what-we-read/plans-and-situated-actions-lucy-suchman/index.md): A plan is not what people execute, but a resource they use to orient themselves while improvising inside a specific situation.
- [The Creators · Daniel J. Boorstin](https://www.3.14financialcontents.com/en/culture-314/what-we-read/the-creators-daniel-j-boorstin/index.md): Boorstin runs through some three thousand years of artistic and intellectual creation and lets a craft show through: creating for an audience and a result, with the means of each age and often ahead of them.
- [The Design of Everyday Things · Donald Norman](https://www.3.14financialcontents.com/en/culture-314/what-we-read/the-design-of-everyday-things-donald-norman/index.md): An object communicates what can be done with it and gives back a signal of what has happened; when one of the two fails, the blame appears to lie with the person using it.
- [The difference between logging an action and observing its result](https://www.3.14financialcontents.com/en/culture-314/notes/the-difference-between-logging-an-action-and-observing-its-result/index.md): A log says the system did something; only an independent signal says that something happened.
- [Thinking in Systems · Donella Meadows](https://www.3.14financialcontents.com/en/culture-314/what-we-read/thinking-in-systems-donella-meadows/index.md): A system's behaviour comes from its structure, and the points where an intervention changes something are rarely the most visible ones.
- [What context does a bot need before acting on an ERP?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-a-bot-needs-to-know-before-acting-on-an-erp/index.md): Most of an agent's work on an ERP happens before the first write, and consists of knowing what it cannot do.
- [What data was actually available when the decision was taken?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-data-was-actually-available-when-the-decision-was-taken/index.md): Reconstructing a decision with today's database produces an unfair judgement and an optimistic model.
- [What does it mean that a bot has finished a task?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-does-it-mean-that-a-bot-has-finished-a-task/index.md): Finishing the execution and producing the expected effect are two different facts, and usually only one of them gets recorded.
- [What does robotics teach about silent software failures?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-robotics-teaches-about-silent-software-failures/index.md): Robotics has spent decades assuming that a component can work and still produce no result at all; business software still discovers it case by case.
- [What must stay invariant through a localisation?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-must-stay-invariant-through-a-localisation/index.md): Localising well requires deciding beforehand what cannot move, and that list is almost never written down.
- [What should a process that may run for months remember?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-should-a-process-that-may-run-for-months-remember/index.md): A long-running process outlives deployments, changes of policy and sometimes the people who started it.
- [What should an organisational memory forget, and what should it keep?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-should-an-organisational-memory-forget-and-what-should-it-keep/index.md): A memory that keeps everything ends up retrieving expired decisions with the same authority as the ones still in force.
- [What we are learning by taking operations into the physical world](https://www.3.14financialcontents.com/en/culture-314/lab/what-we-are-learning-by-taking-operations-into-the-physical-world/index.md): The difference from a digital operation is not the difficulty of the control, but the fact that the state has to be estimated.
- [When does a confidence metric create a certainty the system does not yet have?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/when-a-confidence-metric-produces-too-much-confidence/index.md): A figure that is well calibrated in aggregate can be badly calibrated for exactly the case somebody has in front of them.
- [When does a tabular model deliver more value than a language model?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/when-catboost-is-a-better-answer-than-a-language-model/index.md): On tabular data with labels and a known cost of error, gradient boosting still wins on accuracy, cost and the ability to analyse the result.
- [When should a system recommend, decide, or do nothing but present evidence?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/when-should-a-system-recommend-decide-or-do-nothing-but-present-evidence/index.md): The three stances use the same model and distribute responsibility in three different ways.
- [Which decisions should stay local?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/which-decisions-should-stay-local/index.md): The link drops at the worst possible moment, and whatever is decided then was already decided beforehand.
- [Which part of a narrative should be generative and which deterministic?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/which-part-of-a-narrative-should-be-generative-and-which-deterministic/index.md): The same piece contains data that admits no variation and explanations that need it.
- [Which responsibilities must stay deterministic in an agent system?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-must-stay-deterministic-in-an-agent-system/index.md): The agent can choose the route; permissions, calculations and writes should behave the same way in every execution.
- [Which signal shows that a physical action produced the expected result?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/which-signal-shows-that-a-physical-action-produced-the-expected-result/index.md): The command sent, the actuator moved and the object in place are three facts that often get merged into one.
- [Why a question can be an object of knowledge](https://www.3.14financialcontents.com/en/culture-314/notes/why-a-question-can-be-an-object-of-knowledge/index.md): We publish our open questions with the same structure as a note: why they matter, what we know and what remains unresolved.
- [How can a system infer business logic without turning every historical habit into a rule?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-can-a-system-infer-business-logic-without-turning-every-historical-habit-into-a-rule/index.md): Code, configurations, documents and traces show how the work has been done, but on their own they do not prove which logic ought to be kept.
- [Where should a person step in to contribute judgement without becoming a bottleneck?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/where-should-a-person-step-in-to-contribute-judgement-without-becoming-a-bottleneck/index.md): Human intervention brings knowledge and authority, but an imprecise design can turn it into a queue of reviews without context.
- [How does a system learn from a resolved exception without generalising it too far?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-does-a-system-learn-from-a-resolved-exception-without-over-generalising-it/index.md): An exception contains valuable operational knowledge, but it may depend on a context that disappears when it is turned into a rule too soon.
- [What should a system optimise when time, cost, quality and risk compete with one another?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/what-should-a-system-optimise-when-time-cost-quality-and-risk-compete/index.md): An operation rarely has a single objective; improving one metric can worsen another or shift the cost to a different phase.
- [How is an operation shared between agents, deterministic software and people without losing a single point of responsibility?](https://www.3.14financialcontents.com/en/culture-314/what-we-ask/how-is-an-operation-shared-between-agents-deterministic-software-and-people-without-losing-a-single-responsibility/index.md): Distributing tasks can improve the system, but it can also fragment the context and leave each component considering only its own work finished.

## Other languages

- Español (es): https://www.3.14financialcontents.com/llms.txt · https://www.3.14financialcontents.com/llms-full.txt
- Galego-Português (pt-PT): https://www.3.14financialcontents.com/llms-gl-pt.txt · https://www.3.14financialcontents.com/llms-full-gl-pt.txt