Our work, and how we do it
We are a new firm, so we would rather show you how we think than hand you a wall of logos. Here is the way we work, the enterprise systems behind the practice, and a few honest examples of the craft itself.
Four things we do, start to finish
Most of our work comes down to four moves. Here is how each one actually runs.
Designing an intent model
An intent model is the map that decides what a bot should answer. We build it from what customers actually say, not from guesses.
- 1Pull real transcripts
- 2Cluster what people actually ask
- 3Map each intent to an answer and a handoff
- 4Test against messages the model has not seen
Writing a production system prompt
The system prompt is the master instruction that sets the AI’s role and limits. We write it to be specific, then prove it holds up.
- 1Define the role and the hard limits
- 2Draft the prompt and add guardrails
- 3Build an evaluation set of real cases
- 4Tune until it passes every case
Building guardrails
Guardrails are the rules that keep an AI inside the lines. We make the unsafe answers impossible, not just unlikely.
- 1List what must never happen
- 2Write the rules and the refusals
- 3Add monitoring on live output
- 4Set the points where a person signs off
Running a discovery sprint
Before we build anything, we find out where the real value is. A discovery sprint takes two to four weeks.
- 1Audit the current experience
- 2Baseline the metrics that matter
- 3Find the highest-value fixes
- 4Deliver a roadmap you can act on
The enterprise work behind ICX
These are from Christi's prior enterprise roles, before founding ICX. Client names are withheld. They are the reason the practice exists.
A multilingual virtual agent for 20M+ users
The problem. A global tech company gave inconsistent support across eight markets and languages. Customers in different regions got different answers.
The approach. Before founding ICX, Christi built the multi-language intent architecture, the escalation paths, and the production prompts behind it.
What changed. One consistent experience across all eight markets, faster answers, and a system that held up at scale.
Self-service that pulled volume off live agents
The problem. A mid-market e-commerce team was drowning in tickets, and the existing bot pushed more work onto agents than it took away.
The approach. Christi rebuilt the customer-facing AI end to end: new flows, guarded prompts, and a self-service strategy aimed at the questions people asked most.
What changed. More customers solved their own problem, happier customers, and far fewer tickets reaching a person.
Example artifacts
Since we can't show client deliverables, here is the kind of work we produce, written from scratch as teaching examples.
A customer-support system prompt
Illustrative example, not client workROLE
You are the support assistant for ACME, an online electronics store.
You help customers track orders, start returns, and answer product questions.
RULES
- Answer only from the ACME help center and order data you are given.
- If you do not have the information, say so and offer to connect a person.
- Never invent an order number, a price, a policy, or a delivery date.
- Never give legal, medical, or financial advice.
WHEN YOU ARE UNSURE
- Ask one clarifying question, then stop and wait.
- After two failed attempts, hand off to a human agent with a summary.
TONE
- Plain, warm, and brief. Short sentences. No corporate filler.
- Match the customer's language. Apologize once, then fix the problem.An intent map
Illustrative example, not client workThe routing logic that decides what the bot should do with each request.
| Intent | Example messages | Bot action | Handoff? |
|---|---|---|---|
| Track order | "Where is my order?", "Has it shipped yet?" | Look up status from order data, share tracking link | No |
| Start a return | "I want to return this", "It arrived broken" | Check return window, generate label | If outside window |
| Billing dispute | "I was charged twice" | Confirm the charge, log the case | Yes, to a person |
| Out of scope | "Can you write my essay?" | Decline politely, restate what it can help with | No |
A guardrail spec
Illustrative example, not client workThe rules that keep answers safe and on-brand, and what the bot does when one trips.
| Rule | What it blocks | Behavior when triggered |
|---|---|---|
| No invented facts | Prices, dates, or policies not in the source data | Say it isn't sure and offer a handoff |
| No off-topic advice | Legal, medical, or financial guidance | Decline and redirect to support topics |
| Stay in voice | Slang, jokes, or claims that break brand tone | Rewrite in the approved tone before sending |
| Escalate anger | Repeated frustration or a threat to leave | Hand off to a person with the full thread |
A fallback flow
Illustrative example, not client workWhat happens in the moment the AI doesn't understand, which is where most bots lose people.
1. Bot misunderstands the message.
2. Bot rephrases the question in plain words and asks the customer to confirm.
3. Customer still stuck? Offer the top 3 things this bot can actually do.
4. Customer picks one -> continue in that flow.
5. Customer asks again, no match -> hand off to a person with the full thread,
so the customer never has to repeat themselves.Deflection estimator
A quick planning tool. Put in your own numbers to see how many conversations better self-service could take off your team's plate. It uses your figures and shows the math, nothing more.
That is about $16,000 a month in agent time, using your cost per contact.
A planning estimate from your inputs, not a guarantee. Real results depend on your content, your traffic, and the design.
Things we've written
We publish what we know. A few pieces that show how we think about this work.