Products you sell
Your own SaaS product, app or platform that customers pay for, with AI features where they help.
- SaaS products
- Web and mobile apps
- AI features customers pay for
Example: Codnity CRM, our own product
Custom software and AI products, built faster with AI, with AI inside. Products you sell, platforms and portals, internal tools, and AI added to the software you already use. A first working version takes days to weeks.
Custom software used to be for big companies with big budgets. AI changed how fast software gets built and what it can do once it runs. Your data is what makes it useful.
AI writes a large part of the code, and people review it.
It reads documents, answers from your files and understands speech. The people using it get answers instead of forms.
Everyone has the same AI models. What they are connected to, your data and your process, is what makes them useful.
Four kinds of project. Each one is built faster with AI, and each can have AI inside.
Your own SaaS product, app or platform that customers pay for, with AI features where they help.
Example: Codnity CRM, our own product
Customer and partner portals, marketplaces and booking platforms, where people serve themselves.
For example, a portal where customers check orders and ask questions
The tools your team works in every day, built around how you work instead of the other way round.
Example: the data analytics agent
AI added to the systems you already have, without replacing them. Your team keeps the same tools.
These are the AI functions we build in most often. Each one turns something people used to type, search or check by hand into something the software does.
Invoices, contracts and forms become data the software can use.
An assistant answers questions from your own documents, and shows the source.
Calls and voice notes become text, so the software can act on them.
Emails, replies and summaries, written for a person to check and send.
Amounts, orders or logins that don't fit the pattern get flagged.
Sales, stock and cash, projected from your own history.
AI integration connects a model like Claude or ChatGPT to the systems your team already works in, with only the access it needs. Open a folder to see what AI does in each one.
Call notes, lead summaries and follow-up drafts inside HubSpot, Salesforce, Pipedrive or Dynamics 365.
Invoices read for you, and questions about spend, stock or cash answered from Business Central, SAP, Horizon, Jumis or Xero.
An assistant that answers from your SharePoint or Drive files, drafts emails and sums up meetings.
Reply drafts, ticket routing and answers in Zendesk, Intercom, Slack, WhatsApp or Telegram.
Product texts, order answers and campaign drafts in Shopify, Magento, Stripe, Klaviyo or Mailchimp.
Search across Notion, Confluence and Jira, and answers from your own databases and reports.
We build our own products with the same team and the same way of working. That is how we know what it takes to get AI software into daily use.
Answers and makes phone calls in your customers' language, books meetings and updates the CRM.
Answers questions about your business data in plain language, with the numbers and the source behind each answer.
Deals, contacts and follow-ups in one place, for sales teams that outgrew spreadsheets.
Start small, see it working on your own data, then build the rest.
The honest part: not everything needs AI. If a simple rule does the job, we use a rule. It is cheaper and more reliable.
Short answers to what people ask before an AI software project. Ask Cody for anything else.
AI software development is building custom software and AI products with the help of AI: products you sell, platforms and portals, internal tools, and AI added to the software you already use. AI writes part of the code, so it is built faster, and AI works inside the software, reading documents or answering from your files.
Much faster than before. With AI, a first working version takes days to weeks, where it used to take months. The full scope depends on what you need, and you see it working before you commit to the rest.
AI software development builds something new: a product, a portal or a tool. AI automation takes manual work off your team with workflows and agents inside the systems you already use. See AI automation services.
Yes. We build software products that customers pay for, from the first working version to billing, accounts and support, with AI features where they help, such as answers from documents or forecasts. We build and run our own products the same way, including Codnity CRM.
A customer portal pays off when customers keep emailing or calling for the same things: order status, invoices, documents or bookings. A portal lets them serve themselves, and with AI inside it can answer their questions from your own data and files.
Yes. When a team runs its work on shared spreadsheets, we build the tool around how it works: one place for the data, the steps and the approvals, with access per role. Most internal tools connect to the CRM or ERP the company already has.
The ones we build most often read documents, answer questions from your own files, turn speech into text, write drafts for a person to check, flag what looks unusual and forecast from your history. We pick them by what saves your users time, not by what is new.
AI integration is adding AI to software you already use: an assistant that answers from your documents, AI inside your CRM or ERP, or AI features in your own app. You keep your systems, and the AI works inside them with only the access it needs.
No. If a simple rule does the job, we use a rule, because it is cheaper and more reliable. We use AI where a step needs reading, language or judgement.
Yes. Customer service, voice, sales follow-up and document agents are part of our AI automation services. When an agent is part of a product you sell, we build it into the product.
The data analytics agent is our own product: it answers questions about your business data in plain language, with the numbers and the source behind each answer. It works on a data warehouse we set up and run for you. See the data analytics agent.
You do. The code, the data and the accounts are yours. We can keep running the software for you, or hand it over to your team.
We pick the model for each job, such as Claude, GPT or Gemini, and build the software so the model can be swapped when a better one comes out.
Each part of the software gets only the access it needs, and every action is logged. We document which AI provider processes what and where. Under the commercial terms we use, your data is not used to train AI models.
It depends on the scope. After a short call, we price the first version as a fixed project, so you know the cost before we start.
An idea, a process or a tool your team is missing. We reply within one working day.
New software collects data and holds your customers' details. Make the data useful, and keep it safe.
What we learn building AI systems, reports, secure software and stores, written down.
All insightsCody