Skip to content
PlanBuildAI AI in Construction

AI Agents and Chatbots

AI agents, assistants and chatbots

We build assistants that understand a request, look up the right information in your systems and take the next step — drafting, updating a record, routing a task — with clear limits on what they may do on their own.

Start your project How it works

Covers: AI Agents and Intelligent Assistants · AI Chatbots and Conversational AI

What our team does

An agent is only as useful as the tools it can use and the rules it follows. We define exactly which actions the assistant may take, which need human approval and what it must never do, then connect it to your documents and applications through well-tested integrations. Every action is logged so you can see what it did and why.

  • Task-focused AI agents

    Agents that complete a defined multi-step job — triaging requests, preparing a report, checking a submission against rules — using your tools and data.

  • Internal knowledge assistants

    A chat interface for staff that answers from your procedures, specifications and project records, citing the source for each answer.

  • Customer-facing chatbots

    Website or portal assistants that answer common questions, collect structured enquiries and hand over to your team with the full conversation.

  • Tool and system connections

    Secure, permission-aware connections from the assistant to your CRM, ticketing, document management or project systems.

  • Human-in-the-loop approvals

    Approval steps for actions with consequences, such as sending email, changing records or committing spend.

  • Guardrails and audit logs

    Scope limits, input filtering, rate limits and a complete log of prompts, tool calls and outcomes for review.

Problems we help solve

  • Your team answers the same internal questions every day, pulling them away from higher-value work.
  • Enquiries arrive unstructured and someone has to read, sort and route each one by hand.
  • Knowledge is spread across manuals, past projects and people’s inboxes, and new staff cannot find it.
  • A generic chatbot gave confident but wrong answers, so nobody trusts it.
  • You want to try AI agents but need clear control over what they can access and change.

What you receive

  • A written definition of the agent’s scope, allowed actions and escalation rules
  • The assistant connected to your agreed data sources and systems
  • A web chat interface, an embed for your site, or an integration with your existing tools
  • An evaluation set of realistic questions and tasks, with measured results
  • Admin controls, audit logs and usage reporting
  • Source code, prompts, configuration and documentation

The exact deliverables for your project are listed in your written proposal.

Technologies we use

We choose tools for your requirements, existing systems and the team that will maintain the result — not the other way round. Typical choices include:

  • Claude and OpenAI APIs
  • Tool / function calling
  • Model Context Protocol (MCP)
  • Retrieval-augmented generation
  • Python
  • TypeScript
  • PHP
  • Vector search
  • Webhooks and REST APIs

How the process works

Every AI agent or chatbot project follows the same transparent path. You always know what happens next and nothing is built before you have agreed the scope in writing.

  1. Step 1: Request

    You describe the problem, the systems involved and any constraints in a short form. Attach documents if they help.

  2. Step 2: AI-assisted qualification

    An AI assistant organises the request and flags missing details. It does not price or approve anything.

  3. Step 3: Review by our engineers

    An engineer on our team reads the request, checks feasibility and contacts you with any questions.

  4. Step 4: Written proposal

    You receive scope, deliverables, timeline, price and assumptions in writing — nothing is implied or hidden.

  5. Step 5: Agreement

    Once you accept, we put a formal agreement in place before any build work starts.

  6. Step 6: Delivery

    We build in short, reviewable increments, test against the agreed acceptance criteria and hand over.

Frequently asked questions

A chatbot answers questions in a conversation. An agent can also take actions — looking up records, filling in forms, calling other systems — to complete a task. Many projects start as an assistant and add actions gradually.

We restrict it to a defined set of actions, give it only the access it needs, require human approval for consequential steps and log everything. We also test it against difficult and adversarial cases before launch.

Yes. We index the documents you choose, keep the index updated as they change and make the assistant cite the passages it used, so answers can be checked.

Usually, if those systems offer an API or another supported way to connect. We confirm each integration during the review of your request, before you receive a proposal.

It says so and offers a hand-over to a person, passing along the conversation so far. We design that path deliberately rather than letting the model guess.

Related services

Looking for something else? See all our services.

Start your AI agent or chatbot project

Tell us what you need in a few minutes. An engineer on our team reviews every request and replies with questions or a written proposal — scope, timeline and price included.

Start your project

Or start with: AI Agents and Intelligent Assistants AI Chatbots and Conversational AI

Free and without obligation — work starts only after you accept a written proposal.