EN TR
Get a quote
View all services
Web software

One app carries the brochure pages and the request queue.

System architecture

The pieces and their boundaries are settled before code.

Enterprise & commercial software

Several institutions, one system, records kept apart.

Mobile app

One codebase, both stores: iOS and Android.

Software consulting

We take a running system over and improve it.

AI consulting

AI only where the problem actually calls for it.

Get a quote
EN TR

AI Consulting · Kartal

AI Consulting for Kartal businesses

A new tool does not arrive at a workshop all at once. It goes to one person first, that person runs it for a stretch, and then reports back to the table. If that is how you reach a decision in a business around Cavusoglu and Kordonboyu, the trial itself settles it. What ends a trial is often not what the tool can do; it is the screen opening empty on the second visit. Secretum, which we are writing right now, keeps what accumulates under the account rather than inside the session.

Get a quote Tell us about your project

AI Consulting — Kartal

Kartal leans on the manufacturing strip along the E-5 and the area around the health campus. If a software decision is not made in one sitting, the tool stays in someone's hands for a stretch, and whatever builds up during that stretch settles the argument. The work closest to this district's business profile is our own product Secretum, which is not released and is still being written. Inside it a user builds an AI character that posts into a feed; for production tracking, service calls or clinic bookings we have nothing there to show. MongoDB holds the character record and the Anthropic API writes the text, with Vue 3 on the browser side, React Native (Expo) on the phone and Laravel beneath both.

  • Someone who opened a character days earlier finds the same character on return, because its identity and everything gathered until then live in a durable record under the account.
  • Closing a session is not a reset, so a trial running into a second week still has the first week's build-up in front of it.
  • The limit is written down: Secretum is a social feed product, and production tracking or service call handling sit outside it.

Why Algoriel

  • The image does the reading

    Vipkurum reads optical form marks with OpenCV from the image of the sheet, not through the optics of a reader appliance.

  • Cropping apart from judgement

    Image processing decides where a question begins and ends; the language model only binds the answer to it. The two jobs never mix.

  • Model output lands in a record

    Vipkurum writes the answer key into the cropped question's own record instead of keeping it as a separate document.

  • A person has the last word

    At Tahlil.com the model drafts the reading and a doctor approves what reaches the patient; an unapproved reading never appears on screen.

What we build

AI Consulting

  • Where no model is needed

    Marks on an answer sheet are read from the form image itself; the threshold that decides whether a bubble counts as filled sits in the code, and no language model runs in that step.

  • Splitting the job in two

    Computer vision sets the question boundaries; a language model then ties each answer in the key to its question. Cropping accuracy never rests on a model's interpretation.

  • Removing the hardware step

    Marking stops depending on a reader device; when the form layout changes, we adjust the reading service in software instead of ordering new hardware or device-specific forms.

  • Turning existing material into a pool

    A booklet page is split into individual questions, the answer key is written onto each question's own record, and the questions already on hand become the pattern for new ones.

  • Tying the output to a person

    At Tahlil.com AI prepares the first reading, a specialist physician writes and approves the interpretation, and responsibility for the text that reaches the patient stays with the physician who signed it.

  • Queueing the generation

    In Secretum text and images come out of two separate generation paths; image generation is not instant, so that work goes onto a queue, and the finished file is stored once instead of being regenerated in the feed. Secretum is the product we are building now.

Work like this

AI Consulting · Case study

Secretum — what a character gathers sits under the account, not in the session

Our own product, still being written rather than handed to a client. A user builds an AI character and the character posts into a feed. The split we made: the character's identity and everything it has gathered up to that day do not live inside a chat session but in a durable record held under the account. Come back days later and the character carries on from where it stopped, so a second visit does not begin from nothing. MongoDB holds the record, the Anthropic API produces the text, and Laravel takes the records. The limit is plain: this is a social feed product, not production tracking or service management.

Read the case study →

Working with Kartal teams, remotely

We have no office, so work for this city runs remotely and we do not offer on-site support. What was done, and why it was done that way, stays in writing. The project pages mark which builds came out of a partnership.

  • The first release ships as a working piece; extras follow later.
  • Every client we know of is in Turkey.
  • Decisions stay in writing; progress shows in working screens.

Frequently asked

  • We trial a tool for a stretch and then decide. What happens to whatever builds up during the trial?

    That depends on what the record is attached to. In Secretum a character's identity and gathered history are held under the account rather than in a chat session, so a user coming back after several days finds a full screen instead of an empty one.

  • Can we take Secretum and run it on our own server?

    No. The product is still in the writing stage and we do not describe it as finished. What travels from here is not the product itself but the decision about whether a record hangs off a session or an account.

  • Would you be training a language model for us?

    We train no models; we call ready model interfaces, and in Secretum the Anthropic API produces the text. The harder question is rarely what the model writes. It is where the thing it wrote is kept.

  • Do we not have to be at the same table to run a trial together?

    No. We keep no physical office and the work runs remotely from start to finish. What we discuss during a trial is not the look of a screen but which piece of information lands in which record.

AI Consulting in other cities

The other cities where we publish this service are below, grouped by province.

Other services for this location

Let us look at which step needs a model and which one needs computer vision.

Get a quote or open the service