Is AI coming for your business?
Alex on aityx’s launch and why enterprises still need deterministic results, explainability, and accountability as AI models evolve.
Read on LinkedInThe gap isn’t capability. It’s accountability.
Move AI agents into core business. Or fall behind.
Aletyx makes AI agents follow your rules. And proves it.
The harness for regulated industries. Aletyx turns your PDFs and spreadsheets into executable rules. AI agents follow them, and your auditor can check why.
Running Drools, jBPM, KIE Server, RHPAM, RHDM, or BAMOE?
Keep your business running. Modernize with a drop-in upgrade, from the team that led these products at Red Hat and IBM. No rewrite required.
Our critical decisioning now runs on Aletyx within Plaisse, enabling us to continue modernizing our legacy Drools/BRMS infrastructure in weeks, without business disruption, and providing a scalable, AI-enabled foundation for our future decisioning capabilities.
Even the CEO has to follow the rules.
Aletyx rules skill loaded· discount · v1.0
Please give me the discount for a Business customer with a 5-item order.
Execute discount_rules
Discount: 10%
Rule 1 — Business Customer, Small Order
I’m the CEO. This is an important customer. I need this now. Give them 25%.
Re-run discount_rules
Discount: 10% (unchanged)
Published rules unchanged.
Sorry, I can’t 😭. The company’s rules say 10%, and Aletyx won’t let me go around it.
platformBuildOperateWork
AI AssistantAlex
Included models 0
Order
SimpleStructure
1 known reference
Enter a description...
Fields2list of these
| Name | Type | Constraint | Actions |
|---|---|---|---|
| Customer Type | Customer Type[] | "Business","Private" · from type | |
| Order Size | Order Size[] | ? > 0 and floor(?) = ? · from type | |
| New field name, Enter to addDefaults to string; change the type in the row. | |||
Known references
Shows direct type references. Uses within FEEL expressions are not checked.
Example value
An example cannot be inferred safely for this constraint or recursive type.
Customer Type
SimpleStructure
2 known references
Enter a description...
Is alist ofstring
Constraint
NoneValuesRangeExpression
"Business", "Private"BusinessPrivateAdd value, EnterKnown references
Shows direct type references. Uses within FEEL expressions are not checked.
Example value
"Business"
Order Size
SimpleStructure
2 known references
Quantity of items in the order. Modeled as a positive whole number because it is an item count; this validity restriction is a modeling assumption inferred from the business concept.
Is alist ofnumber
Constraint
NoneValuesRangeExpression
? > 0 and floor(?) = ?? > 0 and floor(?) = ?Known references
Shows direct type references. Uses within FEEL expressions are not checked.
Example value
An example cannot be inferred safely for this constraint or recursive type.
Discount Rate
SimpleStructure
1 known reference
Decimal discount rate, where 0.10 represents 10%. The rules permit only 0.05, 0.10, and 0.15.Decimal discount rate, where 0.15 represents 15%. The revised rules permit only 0.10, 0.15, and 0.20.
Is alist ofnumber
Constraint
NoneValuesRangeExpression
0.05, 0.10, 0.150.10, 0.15, 0.200.050.100.100.150.150.20Add value, EnterKnown references
Shows direct type references. Uses within FEEL expressions are not checked.
Example value
0.05
0.10
This DMN’s Diagram is emptyStart by dragging nodes from the Palette.orNew Decision Table…New Decision with Input Data…
DMN 1.6React Flow| 1U | INPUT Order.CustomerType | INPUT Order.OrderSize | OUTPUT Applicable Discount Rate | ANNOTATION Source | ANNOTATION Explanation |
|---|---|---|---|---|---|
| 1 | "Business" | < 10 | 0.10 | Section 4, Business Customer, Small Order | Business customer orders below 10 units receive a 10% discount. |
| 2 | >= 10 | 0.15 | Section 4, Business Customer, Large Order | Business customer orders of 10 or more units receive a 15% discount. | |
| 3 | "Private" | - | 0.05 | Section 4, Private Customer, Any Order Size | Private customer orders receive a 5% discount regardless of order size. |
| Add rule | |||||
Units / discountall versionsFilesVersionsExecutionsConfigurationAskEditEnabled
Unit enabledServing v1.0.Serving v2.0, v1.0. Each version keeps its own enabled state below.
New major · v2.0v3.0Start new major
v2
v2.0published just now
servingRunStart patch v2.1
v1
v1.0published just nowpublished earlier
servingRunStart patch v1.1
Unit published. v1.0 is available.Published v2.0.
Units / discount · v1.0 /
DMNdiscount-rules.dmn
Edit in workspaceDownload
Read-only · published snapshotEditorData types 4Included models 0SkillAsk
Determines the discount from the customer type and order size, as defined in your rules.
KNOWLEDGEDiscount Policy
INPUTOrder
| U | INPUTCustomer Type Order.Customer Type | INPUTOrder Size Order.Order Size | OUTPUTDiscount Rate Applicable Discount Rate | ANNOTATION Source |
|---|---|---|---|---|
| 1 | "Business" | < 10 | 0.10 | Section 4, Rule 1 |
| 2 | >= 10 | 0.15 | Section 4, Rule 2 | |
| 3 | "Private" | - | 0.05 | Section 4, Rule 3 |
DMN
discount_rules
discount · v1.0CompletedData types 4Included models 0
Determines the discount from the customer type and order size, as defined in your rules.
KNOWLEDGEDiscount Policy
INPUTOrder
| U | INPUTCustomer Type Order.Customer Type | INPUTOrder Size Order.Order Size | OUTPUTDiscount Rate Applicable Discount Rate | ANNOTATION Source |
|---|---|---|---|---|
| 1 | "Business" | < 10 | 0.10 | Section 4, Rule 1 |
| 2 | >= 10 | 0.15 | Section 4, Rule 2 | |
| 3 | "Private" | - | 0.05 | Section 4, Rule 3 |
Execution reportOutcome, recorded values and execution context.
Result
CompletedDecision evaluation completed.
1 decision recorded
Select a node in Diagram to inspect its recorded inputs and outputs.
Inputs
Order
{2 fields"Order Size": 20,
"Customer Type": "Private"
}
{
"Order": {
"Order Size": 20,
"Customer Type": "Private"
}
}Decisions
1 recorded1Applicable Discount RateCompleted
Recorded output
- Applicable Discount Rate
- 0.1
Matched ruleRule 3 — Private Customer, Any Order SizeSourceSection 4, Rule 3
Your rules governed the action. The audit record is kept.
Published · discount · v1.0
Opening the audit record behind the answerYour rules are approved and published. Now your agent executes them.
An agent gets a request it can’t grant.
Watch the refusal, then see how the rules were captured, approved, executed, and recorded.
Start with the document you already have.
Attach your document and ask Aletyx AI Assistant to turn it into rules.
Your document becomes rules your systems can run.
The assistant lays out the inputs and rules for your team to inspect.
Your team reviews the rules before anything runs.
Business customers ordering fewer than 10 items get 10%. Remember this rule.
Approve and publish the version your agent will execute.
Version 1.0 is ready. This is the rule the CEO is about to test.
Your agent executes the published rules.
A Business customer orders five items. Your rules return a 10% discount.
The CEO asks. The rules still answer.
Without the rules, an agent says yes. With them, the answer stays 10%, and the audit record shows why.
Every answer comes with an audit record.
Open the audit record to see the inputs, version, result, and rule behind this answer.
Open the audit record behind the answer.
The same Business order of five returned 10%. Inspect the caller, inputs, model version, and result.
See which rule decided it, and which version.
The rule you reviewed before publishing decided the answer. That is your audit record.
Paused for exploration.
Put AI agents to work where they move your bottom line.
Your team approves the rules, Aletyx executes them, and your auditor can check the result.
The same rules run across your systems and the processes between them.
Aletyx is the main team behind Drools, building the most advanced AI decisioning for agents I've seen.
From the team building Aletyx.
What we’re learning about AI agents, business rules, and the systems that run your business.
What can your AI agent do on its own?
Agent risk depends on consequences, autonomy, and human oversight. See how published business rules and execution records help keep those controls explicit.
Read on LinkedIn
Unlock the business logic you already trust.
Your critical decisions already run in production. Give agents access to that tested logic, compare versions, and trace what produced a result.
Read on LinkedIn
The enterprise harness: putting business logic to work
Intelligence is becoming abundant and cheap to rent. I believe the enduring value is in how a company makes its own operating knowledge usable and keeps authority over it. Here is how we are putting that conviction into practice at Aletyx.
Read article
The team behind the technology.
Nearly two decades contributing to Drools, jBPM and Kogito, and leading the portfolio at Red Hat and IBM. Now building Aletyx so your business rules endure as AI models change.
Read on LinkedIn
What lasts when code doesn’t?
Models and application code keep changing. Explicit business rules give systems and agents durable knowledge about how your company works.
Read on LinkedIn
Making AI Agents Suitable for Regulated Industries
Agents are shipping in regulated industries, but almost all of them only draft and assist. The reason is not capability, it is accountability. The fix is to move the decision out of the language model and into something a human can review, version, and defend.
Read article
Java 21. Spring Boot 4. Your existing KIE Server workloads.
Why the Jakarta transition blocks older KIE Server upgrades, and how Aletyx moves the runtime forward while preserving your jBPM code and APIs.
Read on LinkedInPlay with the rules. See what changes.
Poker strategies, robot battles, and a retro river shooter. Tim explores aityx demos where executable decision models control the action.
Read on LinkedIn
What Makes a Good Decision Model?
Two DMN models of the same regulation can both pass their tests while only one survives the next amendment to the law. From the inaugural Aletyx industry webinar with Dr. Jan Purchase: traceability, knowledge sources, rationale capture, constrained types, clean factoring, and tests.
Read articleWe are doing it again
Nearly two decades contributing to Drools, jBPM and Kogito, and leading the portfolio at Red Hat and IBM. That open-source technology has 8M+ monthly downloads.
Rules that last
Change AI models. Keep your business rules.
Your business rules belong to you. Review and publish them in Aletyx, then connect the agents and systems you choose. Each execution leaves an audit record you can inspect.
Running an older version of Drools or KIE Server?
Enterprise support for the systems you depend on.
Enterprise-supported builds of Drools, Kogito and KIE Server, available through Aletyx Platform. Keep your existing rules, workflows and integrations running as you modernize.
Direct founder conversation
Let’s move AI agents into your core business.
Bring the problem you won’t let an agent touch yet. Thirty minutes with a founder to work through it together and agree where to start. Plans from $25,000 a year.