Enterprise AI ← Back to Home Vision Pillars Tenets Architecture UX Where the money is Field scan Compliance Questions

A living strategy map

Enterprise AI: pillars, tenets, and the platform underneath

The vision: enterprise AI should compound organizational value over time. Successful agentic workflows become discoverable, reliable, reusable building blocks; meet users in the surfaces where they already work; progressively redesign and automate work; and improve through continuously maintained context. Underneath: three technical pillars, seven strategic tenets, one architecture for both enterprise-facing and customer-facing agents.

Created Aug 16, 2026 · Last updated Aug 21, 2026

The vision

From the handwritten notes, Aug 2026. The four commitments in those notes live on below as tenets 1 through 3 and technical pillar 1. What remains here is the shape of the journey: every workflow climbs an autonomy ladder, and every completed climb feeds the loop that makes the next workflow cheaper to build.

The autonomy progression

1Assistanswer · summary · rec 2Acta bounded next step 3Anticipatedetect the triggers 4Automate ritualsrecurring, with oversight 5Compounda reusable building block
Every workflow climbs the same stairs: from producing information for a human, to safely taking the next action, to detecting the stimuli humans act on, to operating the ritual itself, to being captured as an organizational building block. Tenets 4 and 7 govern how fast a given workflow may climb.

The compounding loop

  1. Observe repeated work patterns and collaboration rituals.
  2. Identify the native surface where the work happens.
  3. Decide whether to assist the existing workflow or redesign it.
  4. Build and validate a reliable agentic workflow.
  5. Package it as a discoverable, reusable building block.
  6. Learn from execution and update the relevant contextual knowledge.
  7. Reuse the improved building blocks in additional workflows.

The three technical pillars

What has to exist, regardless of which agents get built on top.

PILLAR 01

Data and context

Identify the structured and unstructured data sources, put a layer of tools above them, and make them accessible to every agent in the organization. The data does not move to the agents; the tool layer brings governed access to the data. The knowledge base is contextualized and personalized, and freshness is a managed property: decide explicitly when knowledge should be updated and when it should decay. An agent that cannot reach the right context at the right depth is an expensive chat window.

structured + unstructured sources · knowledge graph · ontology · semantic layer · ingestion (chunk, vectorize, rank) · tool layer over data · freshness + decay

PILLAR 02

Governance

How do I manage all of these agents? What does my compute look like? What does it all cost? What do safety guardrails and security look like? How does auth work for agents? Governance answers these in code, not in policy documents: agent identity, compute allocation, cost controls, guardrail enforcement, and security boundaries, managed the way a fleet is managed.

agent identity + auth · sandbox · security · guardrails · compute · cost controls · quotas

PILLAR 03

User experience

For developers and users alike. The bar is not "teams can build an agent." The bar is teams composing agentic loops out of agents that already exist, and end users getting surfaces that fit the job. Compound existing value instead of rebuilding it: every new agent should make the next team's agent cheaper to build.

CLI · SDK · APIs · discoverability (agent catalog) · composable agent loops · native surfaces: terminal, email, dashboard, chat, embedded, background

The seven strategic tenets

The direction-setting rules that outlive any single project. Pillars are what has to exist; tenets are how you decide. Tenets 1 through 4 come straight from the vision notes; 5 through 7 are what running agents in production teaches.

TENET 01

Compounded value

The platform is measured by whether value stacks: agents built on agents, tools reused across teams, each investment raising the floor for the next one.

TENET 02

Embedded user experience

Identify the repeatable usage patterns of each function and meet it in its native surface. Developers live in terminals and CLIs, sales and marketing live in email, analytics lives in dashboards. The agent goes to where the work already happens, not the other way around.

TENET 03

Proactive workflow optimization

Redesign the workflow rather than bolting a chatbot on top of it. Ask which triggers, rhythms, and rituals currently cause humans to act, and which of those the agent can own. The test for any copilot feature: does the recruiter need a summary to act on, or can the agent email the candidate directly, at least in the early stages?

TENET 04

Proactivity over reactivity

Agents should need less and less prompting to make the right call. They run the floor and pull humans in at the right moments, rather than humans running the show and pulling agents in. The directionality flips: not humans using agents, but agents using humans, with the human as the API for judgment, taste, and authority.

TENET 05

Auto improvement

The system gets better from its own operation: production outcomes feed back into better agents, better prompts, better routing. Improvement is a loop the platform owns, not a quarterly project.

TENET 06

Continuous evals

Eval is not static. You need a constant process for evaluating the outcomes agents generate, and it plausibly deserves 50 percent of your bandwidth. If half the effort on an agent is not going into knowing whether it works, the agent is running on hope.

TENET 07

Human-in-the-loop intelligence

For every feature, decide how much automation is right by asking what failure costs. Name the risk and its vectors, then place humans accordingly: cheap-to-undo actions run free, expensive or irreversible ones get a human gate, and the dial widens as trust is earned.

Reference architecture

Requests enter through the gateway; agents run inside the sandbox and reach everything through the tool gateway. Unstructured sources flow through chunk → vectorize → rank back into retrieval. Every run emits into the evaluation and observability plane.

Gateway auth · routing · quotas RUNTIME Agent sandbox isolated execution · bounded loops Tool gateway tool registry · per-agent policy TOOLS & DATA Tools APIs · actions · analytics (NL→SQL) Structured sources warehouses · APIs · systems of record Unstructured sources docs · wikis · tickets · transcripts INGESTION Chunk split by structure Vectorize embed + index Rank retrieve + rerank EVALUATION & OBSERVABILITY PLANE Evals continuous · outcome-based Traces every run documented Observability agent-aware telemetry Dashboards TTFT · TPOT · alignment · usefulness
Green marks the load-bearing pieces: the sandbox and the eval loop.

UX: enterprise-facing vs customer-facing

The platform underneath is the same. What changes is who is on the other side of the screen. An employee will forgive a rough edge to get their work done; a customer will not. That single difference drives every row below.

Enterprise-facingCustomer-facing
Who it servesEmployees doing their jobs: analysts, support staff, engineers, opsPaying customers who chose you and can leave
Example surfacesInternal copilots, plain-English analytics (NL→SQL), document processing, background workflow agentsSupport agents, onboarding guides, in-product assistants, voice agents
Speed expectationSeconds are fine for hard questions; minutes are fine for background jobsFirst words on screen in under a second, always streaming
Reliability barMostly right, with the working shown, beats slow and perfect; users can verifyA wrong answer is a broken promise; verification is your job, not the customer's
PolishFunctional wins; rough edges are tolerated if the work gets doneTone, brand voice, and edge-case handling are part of the product
When humans step inReview before acting on numbers; gates at irreversible steps onlySeamless escalation to a person; the customer never feels stuck with a bot
Cost of a failureRework and lost time, contained internallyLost trust, churn, screenshots on social media
What you spend onCoverage: more workflows, more agents, more compoundingReliability and polish: evals, guardrails, latency budgets on fewer surfaces

Where the money is

What the research actually says, in plain words. Sources at the bottom of the section.

19 / 20

AI pilots never show up in the company's financials (MIT, 2025)

+34%

more issues resolved per hour by beginner support agents with AI; veterans gained almost nothing

56%

faster task completion for programmers with an AI assistant, in a controlled trial

bought tools succeeded twice as often as home-built ones (67% vs 33%)

1 in 5

companies have actually redesigned a workflow around AI; redesign is the strongest predictor of profit impact

~5 hrs

of expert work is what a top agent can now finish half the time, doubling every few months

Asking your data questions in plain English (NL → SQL)

This is the analytics automation target: instead of filing a ticket with the data team, anyone asks "how did signups do last quarter in Europe?" and gets an answer. The catch: in classroom demos models score around 90 percent, but on real company warehouses (thousands of columns, messy names) they initially scored 10 to 17 percent. Every system that works in production closed that gap the same way:

51% → 90%

Snowflake's accuracy jump, achieved by giving the model a business dictionary, not a better model

40% → 90%

Pinterest's rate of finding the right table, achieved by documenting tables better

10 → 3 min

time to write a query at Uber, across 1.2M queries a month

MIT GenAI Divide 2025 · Brynjolfsson/Li/Raymond QJE 2025 · Peng et al. 2023 · Anthropic Economic Index 2026 · OpenAI GDPval (arXiv 2510.04374) · METR time horizons · McKinsey State of AI Nov 2025 · Stanford HAI AI Index 2026 · Spider 2.0 (arXiv 2411.07763) · bird-bench.github.io · Snowflake Cortex Analyst · Uber QueryGPT · Pinterest Eng · LinkedIn SQL Bot · Databricks Genie

Field scan

What the major players publish as their enterprise AI frameworks, read against this page.

WhoTheir frameWhat they add beyond this page
DatabricksAgent Bricks (Apr 2026)An AI Gateway doing model routing, fallback, and cost optimization in one choke point; context grounded in catalog metadata, lineage, and data quality (claimed 70 percent accuracy lift over plain RAG); agents inheriting the invoking user's permissions; Lakebase for agent memory; the CLEARS eval rubric.
GoogleGemini Enterprise + A2AAn agent gallery users can browse; a no-code builder tier; A2A as a vendor-neutral agent-to-agent standard (agent cards, task lifecycle, signed trust), now at the Linux Foundation, adopted by Azure and Bedrock.
MicrosoftFrontier Firm + Entra Agent IDAgents as first-class directory identities with a full lifecycle: inventory, classification, human accountability, conditional access, shadow-agent discovery. And the org finding: organizational readiness drives twice the impact of individual capability.
AWSBedrock AgentCore (Jul 2025)Memory as its own service (session plus long-term semantic); a credential token vault for agent identity; browser and code-interpreter as separately governed sandboxes.
AnthropicBuilding Effective AgentsSimplicity-first workflow-vs-agent taxonomy; tool design treated as a craft; guardrails that graduate with autonomy: spend caps first, then auto-review, sandboxing, per-workflow cost caps.
OpenAIAI in the Enterprise + Agents guide"Start with evals" as lesson one (agrees with tenet 3); a named guardrail taxonomy with per-tool risk ratings; explicit human-intervention triggers: failure thresholds and irreversible actions (absorbed into tenet 4).
McKinsey · a16zAgentic mesh + CIO surveyGoverned autonomy with defined autonomy levels; org design (AI council, lighthouse transformations); the procurement reality: 37 percent of CIOs run 5+ models, and price and security now outweigh accuracy in model selection.

Compliance

Compliance is not a document written after the fact; it is evidence the platform generates as a side effect of running. The two external anchors: the EU AI Act, whose high-risk obligations began applying in August 2026 (full phase-in through 2027), and NIST's AI Risk Management Framework (Govern, Map, Measure, Manage), the de facto US anchor. Both converge on the same demands, and the platform already produces most of the answers.

What regulators ask forWhere the platform answers it
Risk classification of every AI systemRisk tier assigned at agent creation, in the registry, driving which gates apply (pillar 2 + tenet 4)
Human oversight proportionate to riskTenet 4: gates placed by cost of failure and risk vectors; escalation paths designed, not improvised
Record-keeping and traceabilityTraces: every run documented in the evaluation and observability plane; the audit trail is a by-product, not a project
Accuracy and robustness evidenceContinuous evals (tenet 3): outcome-based, versioned, running before and after every change
Transparency and disclosureCustomer-facing UX: users know when they are talking to an agent and when a human takes over
Data governancePillar 1's semantic layer plus PII scanning and access inheritance at the gateway; agents never see more than their invoking user
An accountable ownerEvery agent carries a named human owner in the registry, and one accountable body (platform + legal + security) owns the mapping above

EU AI Act timeline: artificialintelligenceact.eu · NIST AI RMF: nist.gov/itl/ai-risk-management-framework

Key design questions

The questions to ask of every candidate workflow, from the same notes.

  1. Which user behaviors and workflows repeat frequently?
  2. In which surface does each user naturally work?
  3. Can the underlying workflow be changed instead of merely augmented?
  4. Which triggers, rhythms, and rituals currently cause humans to act?
  5. What can the agent execute autonomously, and where is human oversight required?
  6. How will agents and workflows be discovered, trusted, shared, and reused?
  7. What context does the agent need, and when should that knowledge be updated or allowed to decay?