QIS-E-100 QISAGENT.COM THE INTELLIGENCE LAYER REV 2026-08-28 · BUILD 44
The Intelligence Layer · The Agent Stack

The Agent Stack.

The infrastructure an autonomous financial agent actually runs on — observability, orchestration, retrieval, lineage, constraints. Every entry is described for what it contributes to governing an agent that touches capital, not for what it does in general. That framing is the whole register: there is no shortage of writing about these tools, and almost none of it asks the question a risk officer asks.

QIS-E-100·30 entries·Mapped to MARQUE

How to read this register.

Entries are published at one of two levels, and the level is stated on every page:

  • Profile — scored against five published criteria, with a defended caveat. Requires that someone has actually used it in anger.
  • Listed — not yet reviewed — structured facts and a governance lens, with no score. Honest about what it is.

The second category exists so the register can be complete without pretending to judgment it has not earned. If you build one of these and want a full profile, walk us through it.

Disclosure

No entry in this register has a commercial relationship with the QIS Ecosystem. Listing is editorial and is never sold. Where that changes, it will be disclosed on the entry itself and in partners.json.

Stack category

Analytics engine

A query engine optimized for analytical workloads — scanning and aggregating large columnar datasets — as distinct from transactional processing. Full definition →

QIS-E-401

In-process analytical database — columnar OLAP without a server.

Listed
Stack category

Constraints

Programmable rules applied to what a system may accept or emit, enforced in code rather than stated in policy, and capable of halting execution. Full definition →

QIS-E-408

Programmable validation and correction applied to model inputs and outputs.

Listed
Stack category

Data quality

The degree to which data is accurate, complete, timely and fit for the specific use being made of it — assessed against declared expectations rather than in the abstract. Full definition →

QIS-E-406

Declarative data validation — expectations as executable, versioned assertions.

Listed
Stack category

Evaluation

The systematic measurement of whether a system produces acceptable outputs against defined criteria, on a held-out or adversarial set, before and after deployment. Full definition →

QIS-E-107

Prompt management and evaluation workflows with human feedback.

Listed
QIS-E-108

Model quality, explainability and testing.

Listed
Stack category

Experiment tracking

The recording of what was run, with which data, code, parameters and result, so that a past experiment can be identified and reproduced. Full definition →

QIS-E-105

Experiment tracking, artefact versioning and model registry.

Listed
QIS-E-106

Experiment management and model production monitoring.

Listed
Stack category

Graph database

A store in which relationships between entities are first-class and indexed, so that traversal across connections is efficient rather than a series of joins. Full definition →

QIS-E-304

Property graph database with mature query language and tooling.

Listed
QIS-E-305

In-memory graph database oriented toward streaming and real-time traversal.

Listed
QIS-E-306

Distributed graph database aimed at deep multi-hop analytics.

Listed
Stack category

Ingestion

The movement of data from a source system into an environment where it can be used, including extraction, transport and initial landing. Full definition →

QIS-E-404

Open-source connector framework for moving data between systems.

Listed
Stack category

Lineage

The recorded history of a data element — where it came from, what was done to it, and by which job — sufficient to trace an output back to its sources. Full definition →

QIS-E-405

An open standard for collecting dataset and job lineage metadata across tools.

Listed
Stack category

Observability

The property that a system's internal state can be understood from the signals it emits — traces, metrics and logs — without modifying it to ask. Full definition →

QIS-E-101

Traces, prompts, and evaluations for LLM applications — the closest thing to a decision trail an agent currently produces.

Listed
QIS-E-102

Production ML observability with drift and performance monitoring across model versions.

Listed
QIS-E-103

Open-source evaluation and drift reporting for ML and LLM systems.

Listed
QIS-E-104

Proxy-based logging, caching and cost tracking for LLM calls.

Listed
Stack category

Orchestration

The layer that coordinates multiple agents or steps toward an outcome — sequencing, branching, retry, and the handling of partial failure. Full definition →

QIS-E-201

Role-based multi-agent orchestration with task delegation between agents.

Listed
QIS-E-202

Conversational multi-agent framework with programmable interaction patterns.

Listed
QIS-E-203

Microsoft's SDK for embedding model orchestration into conventional applications.

Listed
QIS-E-204

Graph-structured agent workflows with explicit state and controllable execution paths.

Listed
QIS-E-205

Type-safe agent framework built on Pydantic's validation model.

Listed
QIS-E-206

Open-source agents that write and execute code in a sandboxed environment.

Listed
QIS-E-207

Programmatic optimization of language model pipelines — compiled prompts rather than hand-written ones.

Listed
QIS-E-208

Durable execution — workflows that survive process failure, with complete replayable event history.

Listed
QIS-E-403

Asset-oriented data orchestration with typed dependencies and materialisation history.

Listed
Stack category

Research environment

The interface in which analysis is written, executed and iterated — notebooks, IDEs and their execution models. Full definition →

QIS-E-407

Reactive Python notebooks stored as plain files, with deterministic execution order.

Listed
Stack category

Retrieval

The selection of relevant material from a corpus and its supply to a model as context, determining what evidence the model sees before it answers. Full definition →

QIS-E-301

Data framework for connecting document corpora to language models.

Listed
QIS-E-302

Composable pipelines for retrieval and question answering.

Listed
Stack category

Transformation

The conversion of raw data into analysis-ready form — cleaning, joining, aggregating, deriving — expressed as code and ideally versioned and tested. Full definition →

QIS-E-402

Version-controlled, tested SQL transformations with generated lineage.

Listed
Stack category

Retrieval by similarity in an embedding space, returning items whose vector representation is closest to a query's. Full definition →

QIS-E-303

Managed vector database for similarity search at scale.

Listed