Custom chatbot development for customers and staff

We build chatbots that answer from your own help content, know when to pass a conversation to a person, and show you every week which questions they could not handle.

Short answer

elitics is a chatbot development company with a senior team in Prishtina, Kosovo. We build custom AI chatbots for your website, customer support desk and internal channels such as Slack or Microsoft Teams. They answer from your knowledge base, hand over to a human with the full conversation, and report what people ask. Rates are €45–65 per engineer-hour, or start with a fixed-price pilot.

per engineer-hour for chatbot builds
€45–65
to your support team with the transcript attached
Handover
from approving engineer profiles to the start of work
~48 h
trial period: not satisfied, you do not pay for it
2 weeks

Chatbots we build

AI chatbot for customer support
Answers questions about orders, deliveries and accounts from your help centre and order system, settles routine requests and opens a ticket in your help desk when a person needs to step in.
Website chatbot for pre-sales
Explains products and plans, captures the visitor’s needs, books a demo into your calendar and passes the details to your CRM for the sales team.
Internal assistant in Slack or Microsoft Teams
Gives staff answers on HR policies, IT how-tos and internal documentation where they already work, respecting who may see which documents.
Multilingual chatbots
Reply in the customer’s language from a single knowledge base, with product terms and tone checked per language by your own team.
In-app help assistant
Sits inside your SaaS product, explains features, walks new users through setup and uses the signed-in user’s context to give specific answers.

What is included in a chatbot project

Customers judge a chatbot by the conversations it gets wrong. The effort goes into the content it answers from, the moment it steps aside, and the reporting that shows where it falls short.

Knowledge base connection
Your help centre, FAQs and product documentation connected as the source of truth, re-synced when articles change, with a report of questions your content does not yet cover.
Human handover
Clear triggers for escalation, such as low confidence, frustration, billing disputes or a direct request, and transfer into your live chat or help desk with a transcript and short summary. Outside opening hours, the bot collects contact details instead.
Guardrails and brand voice
The bot stays on the topics you approve, declines out-of-scope requests and never invents prices or policies. It is tested against deliberately difficult conversations before launch.
Conversation analytics
Dashboards for resolution rate, handover rate, unanswered questions and customer ratings, plus a weekly list of failed conversations to review and fix.

How a project starts

  1. 01

    A 30-minute call

    You describe the product, the deadline and your current team. You talk to an engineering lead, not a salesperson.

  2. 02

    Proposal within 24 hours

    Scope, team, timeline and price in writing, as a fixed price or as a monthly team.

  3. 03

    Work starts in about 48 hours

    Once you approve the engineers, NDA and IP assignment are signed and the team is set up in your repositories and tools.

  4. 04

    Two-week trial

    If you are not satisfied, you do not pay for those two weeks, and anyone who is not the right fit is replaced at no extra cost.

Questions to ask a chatbot development company

Most chatbot proposals read alike. These questions separate a bot that lightens your support queue from one that annoys customers.

  1. 01

    How does the chatbot pass a conversation to our support staff?

    A solid reply names the escalation triggers, the help desk or live chat it connects to, what context travels with the customer, and what the customer sees when nobody is available.

  2. 02

    What does it do when the answer is not in our content?

    It should say it does not know, offer a person, and log the question as a content gap. It should never improvise an answer.

  3. 03

    Do we have to switch help desk or live chat tools?

    Listen for integration through the APIs of the tools you already use, a website widget you control, and native Slack or Teams apps for internal assistants.

  4. 04

    How will we measure whether it is working?

    Expect baseline figures taken before launch, then resolution and handover rates, customer ratings and a regular review of sampled conversations.

  5. 05

    Can our team update answers without a developer?

    Good vendors keep your help centre as the source of truth, so editing an article updates the bot, and give you a simple way to flag poor replies.

Ways to start

Fixed-price chatbot pilot
One channel, an evaluation set made from real past customer questions, and a go/no-go decision before you roll the bot out more widely.
AI pair
An AI engineer alongside a full-stack engineer who builds the chat widget, the help desk integration and the admin views your support leads use.
Dedicated squad
From €15k a month, when you need several channels and languages and ongoing tuning based on what customers ask.

Ways to work with us

Staff augmentation

Senior engineers who join your stand-ups and work in your tools.

Best when

You have a roadmap and need more capacity or a specific skill.

Dedicated squad

A cross-functional team (product, engineering, QA, DevOps) that owns a product or feature.

Best when

You want a product built or rebuilt without growing headcount.

Build-Operate-Transfer

Your own development center in Kosovo, set up and run by us.

Best when

You plan a long-term engineering presence in Europe.

See the full service: AI App Development Company

Common questions

How much does custom chatbot development cost?

Our engineers cost €45–65 per engineer-hour. A pilot on one channel is a fixed price agreed after scoping. Running cost depends on conversation volume and the model chosen, and we estimate it per conversation during the pilot.

What happens to customer conversations and personal data?

We use provider settings that keep conversations out of model training, store transcripts in your environment, and mask details such as email addresses and order numbers in logs. An NDA and IP assignment come before any work.

Which AI models power your chatbots?

Typically Gemini or the OpenAI API, often with a smaller, cheaper model for routing simple questions. We use the provider your company has approved and can change models later without rebuilding the bot.

Can one chatbot cover several languages?

Yes. Modern language models handle many languages from one knowledge base. Each language you launch in gets its own test conversations, and your team reviews product terms and tone before it goes live.

Can the same chatbot run on our website and in Microsoft Teams?

Yes. One backend holds the knowledge, rules and logs, and each channel, whether a website widget, Slack, Teams or your app, connects to it with its own permissions.

Let’s build what comes next.

Tell us what you need to build or scale. You will get a proposal and estimate within 24 hours.

Or write to us directly

hello@elitics.io+383 49 171 069

Dukagjini Center, Prishtina, Kosovo