Frequently Asked Questions
Unitic Technology ·
Answers to what we are asked most before a project starts. If yours is not here, write to us from the contact page.
What exactly does Unitic Technology do?
We build systems where off-the-shelf stops — where AI meets mission-critical business processes: enterprise memory and RAG, AI agent systems, LLM security, provable AI, voice and video AI assistants, identity, document and face recognition, network-level protection, multi-tenant SaaS engineering, enterprise platform engineering, plus blockchain and smart contract solutions. Most projects draw on several of these at once.
Which industries do you work in?
We are not tied to one. We have built systems for health and nutrition, the public and cultural sector, education and examinations, a university entrepreneurship ecosystem, digital health, real estate and fitness studios — the Case Studies section shows which work was done where. What they share is not a sector but that an off-the-shelf product would not have fit: places where AI, sensitive data and mission-critical processes have to work together.
How does a project run?
Five stages, the same on every project: discovery and strategic alignment, solution and project planning, delivery and execution, launch and go-live, then support and continuous improvement. Discovery establishes the current state, the business need and the expected business value through a business case, gap analysis and stakeholder analysis; planning turns those into a prioritised backlog, a roadmap and a release plan.
Do you work agile or waterfall?
Either. Planning decides whether a predictive, agile or hybrid management approach fits the nature of the project. The model is chosen for the work; the work is not forced into a model.
What happens after go-live?
Go-live is not the finish line. We manage it together with technical readiness, user acceptance testing, data migration, training, and a rollback and cut-over plan, followed by a hypercare period. Under a maintenance agreement, support, defect management, SLA tracking, performance monitoring and continuous improvement carry on.
What do you build on technically?
Less on a particular language or framework than on how the software gets made. Parallel teams of AI agents write the code; every piece of work goes through a spec, a plan and independent review, and no one approves their own work. Every branch is a release stage, and nothing moves to production past a red quality gate. Client data is separated at the database level, every line traces back to the request it came from, and at the end of the project the rules, decisions and plans become a handbook your team can take over. The Technical Capabilities section covers each of these, along with smart contract engineering, one by one.
Can I see your past work?
Yes — the case studies section covers work such as Hardiye, a 24/7 Turkish-speaking voice and video assistant for Hardline Nutrition; Motice, our own product; an exam and certificate platform and a web network spanning 69 countries for the Yunus Emre Institute; Mentorual, a mentoring platform for İTÜ Ginova; BGB, a protection app against gambling addiction; and more. Where naming a client is not appropriate, the project is published anonymised.
What do you do on the AI side?
Both in client projects and in our own products. The AI Lab section explains how: an LLM security layer that never leaks data to the model, assistants that cannot exceed their user's permissions, identity verification anchored in MRZ check digits, agent memory that knows when something was true, and testing voice assistants with AI. We also build the code itself with AI agents behind human approval gates — that is covered in the Technical Capabilities section.
What is the relationship between Unitic Technology and unitic.co?
Unitic Technology is a subsidiary of unitic.co.
How do we start?
Tell us briefly about your project from the contact page and we will get back to you. You can send it as a general enquiry, a demo request or a technical conversation.