Chandler · usage
← Back to sessions

Session

User
detelina.vassileva@dext.com
Started
2026-08-06 09:15
Last activity
2026-08-06 10:20
Questions
18
Grouping
backfill — reconstructed from timing by migration 011
session_id
ad49e7cb-005d-4d86-8715-3bc5166218fc

Transcript

  1. 2026-08-06 09:15 ask success $0.0196 11.4 s full detail →
    Which fields in our Omni model represent each of these account signals? For each, give me the exact field path (model.view.field) and what value(s) indicate "true": 1. Whether an account is on a plan that includes Core Agents / AI Assist by default (i.e. Core Agents license present) 2. Whether an account has ever started an AI Assist trial (trial start date, trial active/expired flag) 3. Whether an account is currently a paid AI Assist customer (has a paid AI Assist add-on) 4. Whether an account has ever used LIX free credits (LIX free credits consumed > 0 at any point) 5. Whether an account currently purchases LIX bundles (active LIX bundle purchase) Don't return data rows — I only need the field paths and the truthiness rule for each.
    No answer recorded.
  2. 2026-08-06 09:20 ask success $0.8401 256.4 s full detail →
    Cohort: accounts that currently have a paid AI Assist / Agents add-on (paid AI Assist customers), excluding Dext demos and Reseller accounts. Return one row per account with: account CRN, account name, Salesforce country. No aggregation.
    No answer recorded.
    Bookings
  3. 2026-08-06 09:21 ask success $0.3310 60.5 s full detail →
    How many active accounts today have an AI Assist / Agents add-on (i.e. ai_assist_client_limit > 0)? Exclude Dext demos and Reseller accounts. I need the count. Please execute the query.
    No answer recorded.
    1 row Bookings
  4. 2026-08-06 09:24 ask success $1.0877 182.5 s full detail →
    How many accounts have ever started an AI Assist trial (i.e. have an AI Assist trial start date, or an ai_assist_trial_started flag = true)? Please execute the query. Exclude Dext demos and Reseller accounts.
    No answer recorded.
    EXPERIMENTAL Subscriptions Signals
  5. 2026-08-06 09:25 ask success $0.0236 10.5 s full detail →
    Do we track AI Assist / Agents trials as a distinct concept from paid AI Assist? For example: is there an ai_assist_trial_active flag, an ai_assist_trial_start_date, a subscription tier that means "on trial", or a license type value like "trial"? I need to know the field name and how to identify accounts currently on an AI Assist trial vs accounts that have ever started one. Please describe the concept — I don't need data rows, just the field(s) and their values.
    No answer recorded.
  6. 2026-08-06 09:25 ask success $0.0195 9.3 s full detail →
    Two related questions on our booking / subscription model: 1) Is there any way at all to identify accounts currently on an AI Assist trial vs paid, even indirectly? For example: a subscription_tier value, a license_type value, an is_trial flag on the AI Assist booking, or ai_assist_contract_client_limit > 0 combined with something like has_paid_subscription = false. Just tell me the field/logic — no data rows needed. 2) Is there a "Core Agents license" concept? I.e. plans that include Agents / AI Assist by default (not as a paid add-on) — some product feature bundle flag on the plan or account.
    No answer recorded.
  7. 2026-08-06 09:25 ask success $0.0124 7.7 s full detail →
    Which fields in our model track LIX usage at the account level? Specifically: LIX credits consumed (lifetime, or in a period), whether an account has ever used free LIX credits, whether an account currently has an active LIX bundle purchase. I need field paths only. If there's a topic dedicated to LIX / credit consumption, name it.
    No answer recorded.
  8. 2026-08-06 09:31 ask success $1.0074 330.1 s full detail →
    Please execute a query: how many accounts consumed at least one LIX credit (Line Item Extraction) in the last 90 days? Just give me the count. If LIX is not in the model, say so plainly.
    No answer recorded.
    Product Usage - Accounts
  9. 2026-08-06 09:41 ask success $0.7459 213.4 s full detail →
    Please execute this query. Return one row per account: accounts that currently have a paid AI Assist / Agents add-on (ai_assist_contract_client_limit > 0) AND currently purchase a LIX bundle (i.e. have an active LIX bundle subscription today). Columns per row: account_crn, account_name, salesforce_country. Exclude Dext demos, exclude Reseller accounts, exclude suspended accounts. Use the latest available snapshot date.
    No answer recorded.
    Bookings
  10. 2026-08-06 09:42 ask success $0.2129 79.9 s full detail →
    Give me the list of accounts today where BOTH conditions hold: eds_bookings.ai_assist_contract_client_limit > 0 AND the account has a currently-active LIX bundle purchase (any positive LIX bundle MRR or LIX bundle contract value on the latest snapshot). Return account_crn, account_name, salesforce_country. Exclude dext demos, resellers, suspended accounts. Use is_latest_available_date = true.
    No answer recorded.
  11. 2026-08-06 09:44 ask success $0.2027 71.9 s full detail →
    Count of accounts today where ai_assist_contract_client_limit > 0 AND there is at least one active LIX bundle. Exclude demos, resellers, suspended. Just the count.
    No answer recorded.
  12. 2026-08-06 09:46 ask success $0.3915 95.2 s full detail →
    How many paid AI Assist customers are also currently buying LIX bundles?
    No answer recorded.
    1 row Bookings
  13. 2026-08-06 09:48 ask success $0.7304 140.1 s full detail →
    How many paid AI Assist customers have never used any LIX (Line Item Extraction) lines — i.e. lifetime LIX line volume is zero?
    No answer recorded.
    Product Usage - Accounts
  14. 2026-08-06 09:54 ask success $1.1971 341.9 s full detail →
    Among current paid AI Assist customers, how many have zero total LIX line volume?
    No answer recorded.
    Product Usage - Accounts
  15. 2026-08-06 10:00 ask success $1.1230 341.5 s full detail →
    How many paid AI Assist customers have ever extracted any LIX lines (total_lix_line_volume > 0)? Then also give me: of paid AI Assist customers, how many have total_lix_line_volume = 0? Two counts.
    No answer recorded.
    2 rows Product Usage - Accounts
  16. 2026-08-06 10:15 generate internal_error $0.7379 162.2 s full detail →
    Generate a query that returns account_crn, account_name, salesforce_country for accounts where eds_bookings.ai_assist_contract_client_limit > 0 AND account_lix_metrics_dynamic.total_lix_line_volume = 0 AND eds_bookings.lix_contract_bundle_applied = false. Use is_latest_available_date = true. Exclude is_dext_demo = true, exclude salesforce_account_type = Reseller, exclude is_suspended_dext = true.
    Error calling tool 'generate': Failed to produce a valid Omni query after 16 iterations. Active topic: Analytics/Product Usage - Accounts. Last error: blocked: field(s) omni_dbt__eds_all_accounts.account_name contain personal information (name/email/phone/address) and cannot be queried
    Product Usage - Accounts
  17. 2026-08-06 10:16 ask success $0.3095 56.1 s full detail →
    Return distinct account_crn and salesforce_country for accounts where ai_assist_contract_client_limit > 0 AND lix_contract_bundle_applied = true, using is_latest_available_date = true, excluding is_dext_demo, salesforce_account_type = Reseller, and is_suspended_dext. One row per account. Sort by salesforce_country then account_crn.
    No answer recorded.
    53 rows Bookings
  18. 2026-08-06 10:20 ask success $0.9091 220.6 s full detail →
    Return distinct account_crn and salesforce_country for accounts where ai_assist_contract_client_limit > 0 AND lix_contract_bundle_applied = false AND account_lix_metrics_dynamic.total_lix_line_volume = 0. Use is_latest_available_date = true. Exclude is_dext_demo, salesforce_account_type = Reseller, is_suspended_dext. One row per account. Sort by country then CRN.
    No answer recorded.
    Product Usage - Accounts

Cost is the LLM completion spend LiteLLM priced for each call, summed per request. Chandler is running on the Codex CLI, which draws ChatGPT plan quota rather than per-token API billing, so this figure is not money paid — it is what the same traffic would have cost on the API, priced from the token counts Codex reports. Read it as the size of the bill avoided. Two known limits: embedding spend is not recorded, so retrieval and matching cost is missing, and because the figure is one sum per request it cannot be split by model within a request — a request's classifier and agent calls can use different models while llm_model holds only one name. Only authenticated calls are logged, USAGE_LOG_ENABLED can switch logging off, and log writes are fail-soft — this is not a complete record of traffic. Questions are grouped into sessions: a session is exact when the caller echoed its id back to us and otherwise inferred from a 30-minute gap in that user’s activity, so a grouping is only as good as the source shown on the session itself. A call with no attributable user gets no session at all; those questions are listed separately rather than dropped.