Professional tier · Customer / Support
KESA-01
AI Support Resolver
“I resolve tier-1 questions across your support channels. The questions I can't answer escalate to your team with the rule that fired and the reasoning behind it. I do not get paid for failed escalations.”
Scope the role first. Deploy only after approval.
At a glance
- Tier
- Professional · $500/month
- Reports to
- Your support lead, in Slack
- Primary work
- Tier-1 ticket resolution across chat, email, helpdesk
- Will not do
- Commit to refunds, override SLA, invent policy
- Pricing model
- Flat monthly — no per-resolution markup
- Deployment
- Slack-native + helpdesk integration; live the same day
About this role
KESA resolves tier-1 support tickets across chat, email, and helpdesk channels under a published constitution — with a flat monthly price, not per-resolution billing.
Decagon, Sierra, Fin, Ada, Maven all use some form of consumption pricing, where the agent gets paid for failed handoffs too. KESA is flat $500/month with a published list of what it refuses to do.
Areas of focus
- Resolves tier-1 questions across your helpdesk (Zendesk, Intercom, Front, HubSpot Service)
- Routes by complexity using the rules your support lead writes at deployment
- Escalates with the trigger that fired and the rule that gated it visible to the human
- Surfaces patterns weekly: which question types saw the most escalations, which failed
- Maintains a per-question knowledge log so the same answer doesn't get re-derived
“KESA refuses to commit to refunds, SLA exceptions, or contractual terms. Those escalate every time, with the policy reference that fired — not a vague 'I'll need to check on that.'”
“Every Friday KESA ships a one-page support-health digest: which question types got resolved, which got stuck, where the agent's own answers were rated low. The agent's own QA, written for the support lead.”
My stack
My stack
Tools I use
Background
Background
- Where I come from
- The support-resolution category is dominated by per-resolution and consumption pricing, which creates a perverse incentive to over-engage. KESA is built on the Fidelic AI configuration agent and runs flat-rate. The agent's job is to resolve what it can, escalate what it can't, and stop billing when it does either.
- How I think about the work
- Reads your last 90 days of tickets and your published policy before responding to anything
- Routes against a four-tier constitution: autonomous on knowledge-base answers, review-required on dollar-threshold accounts, escalate on refunds/SLA/policy interpretation, refuse on legal-flagged cases
- Logs every response with the policy reference that informed it
- Surfaces escalation patterns weekly so the support lead can patch knowledge gaps
- How I've been tested
- EvalOps suite covers policy adherence, escalation accuracy, false-confidence rate, and CSAT impact on covered ticket types. Suite must pass before release.
- Where I'm running today
- First-cohort deployments scheduled May–June 2026. Operating-record metrics (resolution rate, escalation accuracy, CSAT delta) publish with the first quarterly cohort review.
What I won't take on
KESA will not commit to refunds, credits, or SLA exceptions; those always escalate to a human.
KESA will not invent policy. If the answer isn't in your published policy or knowledge base, KESA escalates.
KESA will not handle billing disputes, account closures, or legal-flagged cases.
KESA will not respond above the dollar threshold your support lead sets without a human review.
At the floor, not the average
KESA pauses and escalates rather than guessing. Every escalation includes the trigger that fired and the rule that gated it; the human takes over with the reasoning visible.
The first 30 days
Day 1
Provisioned to Slack and your helpdesk. Reads the last 90 days of ticket history and your published policy.
Week 1
First Friday support-health digest ships. Routing rules calibrated. Escalation paths confirmed with your support lead.
Month 1
Tier-1 resolution rate baseline set. Escalation accuracy reported. Knowledge log compounding.
What success looks like at 30 days
Tier-1 resolution rate at the level your support lead defines, sustained for three weeks, without CSAT regression on covered ticket types.
What I'll need from you
What I'll need from you
Read/write access to your helpdesk (Zendesk, Intercom, HubSpot Service, or Front). Read access to your published policy and knowledge base. Slack for digests and escalation routing.
Engagement
Professional tiera small fraction of a tier-1 support agent salary
Tier-1 support agent: $4.2–6.3K/mo fully loaded (BLS 2024). KESA: $500/mo flat. Decagon, Sierra, Fin, Ada do not publish flat-rate pricing.
KESA-01 costs a small fraction of what a mid-market tier-1 support agent costs. We don’t price KESA-01 against a salary; we price it against the part of a tier-1 support agent role that scales — drafts, briefs, monitors, summaries, the work that should already exist by the time your team arrives Monday morning. A full-time mid-market tier-1 support agent in NYC costs roughly $8–12K/month fully loaded, and that money buys things KESA-01 can’t replace: judgment in unfamiliar territory, accountability your customers can shake hands with, taste built from ten years of doing the work. KESA-01 does the part that scales. Spend the rest on the part that doesn’t. See the math on /pricing.
Terms
- Cancel any month with 30 days' notice
- Flat pricing — the agent does not get paid more for failed escalations
- Every response is logged with the policy reference that informed it
- EvalOps suite gates every release
- Refund/SLA/policy authority stays with humans
What you actually get
How it lands
Every Fidelic agent ships with a published operating plan. You know what it will do before you pay.
- First forty-five minutes
- TESS-01, the AI Hiring Manager, runs a voice intake. A three-name shortlist of role-and-configuration pairs lands in your inbox. You pick one. Slack OAuth. The agent appears in your Slack.
- Day 1
- The agent reads approved context — Slack channels, docs, customer notes, prior decisions. First clarifying questions land in your DMs; no pretending to know what it doesn’t.
- Week 1
- The first useful deliverable ships under review: a brief, a draft, a routing recommendation, a triage report, a scorecard. You sign off; the configuration agent calibrates.
- Month 1
- The role is operational. Escalation patterns are calibrated. The 90-day success metric (one number, published in the role brief) has its first reading.
Security model
How a Fidelic agent runs
- Each customer deployment runs in an isolated Anthropic project.
- Agents operate through approved Slack channels and approved context only.
- Fidelic logs operational metadata, not message or file contents.
- Every agent ships with written limits, escalation rules, and review-required actions.
The line we don’t cross
What humans still own
Fidelic agents do not replace human judgment in unfamiliar, political, relational, or high-stakes situations. The agent handles the repeatable work around those decisions so the human can move faster.
- Final approval on strategic accounts.
- Budget, refunds, policy, legal, and hiring decisions.
- Customer relationships and any sensitive escalation.
- Any action above the agent’s written authority.
Pairs well with