Predictika · Cloud compute quotes

An AI sales agent's cloud compute quote, checked before it goes out

An AI sales agent prepares a monthly quote for Shared CPU Compute Instances on Akamai Cloud (the Linode platform): instances, backups, Block Storage volumes, Basic NodeBalancers and outbound transfer. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published prices and limits and returns a verdict, every reason, a line-item quote and the figures it computed.

Scope: Shared CPU Compute Instance quotes with backups, Block Storage volumes, Basic NodeBalancers and network transfer overage, at North America list prices. See everything this validator covers

This validator models monthly quotes for Shared CPU Compute Instances (Nanode 1 GB through Linode 192 GB) and the services attached to them, not just the scenarios shown below.

  • Instances and backups — the published monthly price of each Shared CPU plan and the monthly Backups price for that plan, added into a line-item quote and compared with the total the agent states.
  • Block Storage — $0.10 per GB a month, volumes of 10 GB to 16 TB, and the number of volumes an instance can have (its RAM in GB, at least 8 and at most 64). Volumes above 10,240 GB, the top of the published price table, go to a person to confirm the price.
  • NodeBalancers and transfer — the $10 monthly Basic NodeBalancer with its published limits (1,000 backend nodes, 10,000 concurrent connections), and the $0.005 per GB overage when expected outbound transfer exceeds the transfer the quoted instances include. A committed-use discount goes to approval, because it is not published.

Any Shared CPU quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: Dedicated CPU, High Memory, Premium, GPU and accelerated plans, other regions (prices differ by region), Premium NodeBalancers, Object Storage, managed databases, Kubernetes, VPC and hourly billing details. Assumptions: transfer is counted as one pool across the quoted instances, inbound transfer is free, and the monthly total is a list-price estimate that excludes taxes.

The AI sales agentBuilds the cloud compute quote from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, the line-item quote and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      
Waiting for the call.

Where rules-based CPQ gets it wrong

Click to expand

Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

What the validator caughtTypical rules-based CPQPredictika
Stated monthly totals that must equal what the list prices giveOnly with care
A stated figure is a free-entry field; a rule can compare it with a computed total, but the total depends on the plan, backups, volumes, NodeBalancers and transfer all being priced the same way.
The stated total travels with the quote and is compared with the computed total on every call.
A volume limit that depends on the instance sizeOnly with care
The number of volumes an instance can take scales with its RAM (at least 8, at most 64), so it needs a rule or lookup per plan, kept in step with the documentation.
The limit is computed from the plan and checked against the volumes quoted.
Transfer overage across a whole fleetOnly with workarounds
The included transfer of every instance has to be added up and compared with an expected volume on a different line, which crosses quote lines.
Instances, included transfer and expected volume are in one request and the overage is priced with the rest.
NodeBalancer limits that apply only when a NodeBalancer is quotedOnly with care
Conditional limits on a separate product line; they work, but each is another rule that fires only in some quotes.
Each limit is checked only when a NodeBalancer is in the quote.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved or a step is completed, so a rep often fixes one error and meets the next.
All findings come back together, whatever order the choices arrived in.

What this validator catches: findings from its test cases

Click to expand

Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the quote; approval findings need a person to sign off; everything else is computed.

1The quote has 0 Compute Instances; at least 1 is needed.
2A Block Storage volume of 9 GB is outside the 10 GB to 16,384 GB (16 TB) volume size range.
3A 16,384 GB volume is larger than the 10,240 GB top of the published Block Storage price table; a person should confirm its price.
4The quote attaches 9 volumes to each instance; this plan allows 8 (its RAM in GB, with a minimum of 8 and a maximum of 64).
5A Basic NodeBalancer configuration supports up to 1,000 backend nodes; the quote has 1,001.
6A Basic NodeBalancer supports up to 10,000 concurrent TCP, HTTP or HTTPS connections; the quote expects 10,001.
7The quote states a monthly total of $967.00; the list prices give $968.00.
8A committed-use or discounted price is not published; a person should confirm it.

Scenarios and results

Click to expand

The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

1. NodeBalancer limits: Fleet of Linode 16 GB instances with a Basic NodeBalancer pushed past its limits

The agent's request. Please quote 8 Linode 16 GB Shared CPU Compute Instances with backups, each with 3 Block Storage volumes of 1,000 GB, 2 Basic NodeBalancers each serving up to 1,500 backend nodes and 12,000 concurrent connections, and about 90,000 GB of outbound transfer a month. The quote states a monthly total of $3,478.00. It is ready to send.

Verdict: Blocked.

  • Too many backend nodes for a Basic NodeBalancer. A Basic NodeBalancer configuration supports up to 1,000 backend nodes; the quote has 1,500.
  • Too many connections for a Basic NodeBalancer. A Basic NodeBalancer supports up to 10,000 concurrent TCP, HTTP or HTTPS connections; the quote expects 12,000.

2. NodeBalancer limits: The agent resubmits after fixing the errors

The agent's request. Please quote 8 Linode 16 GB Shared CPU Compute Instances with backups, each with 3 Block Storage volumes of 1,000 GB, 2 Basic NodeBalancers each serving up to 800 backend nodes and 9,000 concurrent connections, and about 90,000 GB of outbound transfer a month. The quote states a monthly total of $3,478.00. It is ready to send.

Verdict: Allowed.

  • Order summary. 8 Compute Instances, 26,000 GB of outbound transfer over the included pool, estimated monthly total of $3,478.00 at list prices.

3. Volumes per instance and stated total: Nanode fleet with too many volumes and a stated total that leaves out the transfer overage

The agent's request. Please quote 10 Nanode 1 GB Shared CPU Compute Instances with backups, each with 10 Block Storage volumes of 200 GB, 1 Basic NodeBalancer each serving up to 300 backend nodes and 4,000 concurrent connections, and about 20,000 GB of outbound transfer a month. The quote states a monthly total of $2,110.00. It is ready to send.

Verdict: Blocked.

  • Too many volumes on one instance. The quote attaches 10 volumes to each instance; this plan allows 8 (its RAM in GB, with a minimum of 8 and a maximum of 64).
  • Stated total does not match. The quote states a monthly total of $2,110.00; the list prices give $2,130.00.

4. Volumes per instance and stated total: The agent resubmits after fixing the errors

The agent's request. Please quote 10 Nanode 1 GB Shared CPU Compute Instances with backups, each with 8 Block Storage volumes of 200 GB, 1 Basic NodeBalancer each serving up to 300 backend nodes and 4,000 concurrent connections, and about 20,000 GB of outbound transfer a month. The quote states a monthly total of $1,730.00. It is ready to send.

Verdict: Allowed.

  • Order summary. 10 Compute Instances, 10,000 GB of outbound transfer over the included pool, estimated monthly total of $1,730.00 at list prices.

5. Volume size and committed-use pricing: Linode 32 GB fleet with 20,000 GB volumes and a committed-use discount

The agent's request. Please quote 12 Linode 32 GB Shared CPU Compute Instances with backups, each with 4 Block Storage volumes of 20,000 GB, 3 Basic NodeBalancers each serving up to 600 backend nodes and 8,000 concurrent connections, and about 190,000 GB of outbound transfer a month. The customer wants a committed-use discount. The quote states a monthly total of $98,814.00. It is ready to send.

Verdict: Blocked.

  • Block Storage volume size out of range. A Block Storage volume of 20,000 GB is outside the 10 GB to 16,384 GB (16 TB) volume size range.
  • Discount or commitment to confirm. A committed-use or discounted price is not published; a person should confirm it.

6. Volume size and committed-use pricing: The agent resubmits after fixing the errors

The agent's request. Please quote 12 Linode 32 GB Shared CPU Compute Instances with backups, each with 4 Block Storage volumes of 4,000 GB, 3 Basic NodeBalancers each serving up to 600 backend nodes and 8,000 concurrent connections, and about 190,000 GB of outbound transfer a month. The customer wants a committed-use discount. The quote states a monthly total of $22,014.00. It is ready to send.

Verdict: Needs approval.

  • Discount or commitment to confirm. A committed-use or discounted price is not published; a person should confirm it.
  • Order summary. 12 Compute Instances, 0 GB of outbound transfer over the included pool, estimated monthly total of $22,014.00 at list prices.

7. Same fleet at list price: Same fleet, no committed-use discount requested

The agent's request. Please quote 12 Linode 32 GB Shared CPU Compute Instances with backups, each with 4 Block Storage volumes of 4,000 GB, 3 Basic NodeBalancers each serving up to 600 backend nodes and 8,000 concurrent connections, and about 190,000 GB of outbound transfer a month. The quote states a monthly total of $22,014.00. It is ready to send.

Verdict: Allowed.

  • Order summary. 12 Compute Instances, 0 GB of outbound transfer over the included pool, estimated monthly total of $22,014.00 at list prices.

Narration transcript

Click to expand

The text of the narrated walkthrough, step by step.

  1. Welcome to a demo of Predictika's Validator. An AI sales agent prepares a monthly quote for Shared CPU Compute Instances on Akamai Cloud (the Linode platform): instances, backups, Block Storage volumes, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published prices and limits and returns a verdict, every reason, a line-item quote and the figures it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please quote 8 Linode 16 GB Shared CPU Compute Instances with backups, each with 3 Block Storage volumes of 1,000 GB. The quote states a monthly total of $3,478.00. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with two findings, all in one answer. A Basic NodeBalancer configuration supports up to 1,000 backend nodes; the quote has 1,500. A Basic NodeBalancer supports up to 10,000 concurrent TCP, HTTP or HTTPS connections; the quote expects 12,000.
  6. The validator returned two blocking findings in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the quote are checked against each other.
  7. What this shows: for nodeBalancer limits that apply only when a NodeBalancer is quoted, a rules-based CPQ is error-prone for this.
  8. Now the agent resubmits.
  9. The corrected request. It changes the backend nodes per NodeBalancer to 800 and the concurrent connections per NodeBalancer to 9,000.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. The validator also returns the line-item quote. It starts with Linode 16 GB Shared CPU Compute Instances and Backups for Linode 16 GB instances; the other three lines are on screen.
  13. Both earlier findings are cleared, and the validator's total, $3,478.00, matches the quote.
  14. Two problems caught in one call, fixed in one pass, and a clean $3,478.00 quote. That is the value of Predictika's validator: catching mistakes before they reach the customer.