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 · Uskudar

AI Consulting for Uskudar businesses

Years of booklets sit in the archive cabinet and nobody opens it. If a new worksheet in a school or course building between Kucuk Camlica and Ferah gets written without anyone walking to that cabinet, the reason is plain: a booklet is stored whole, and reaching one question means sending a page to the copier. Material already on hand goes unused and is rewritten instead. Vipkurum's question cropping tool was written for exactly that knot: a booklet becomes a pool of single questions.

Get a quote Tell us about your project

AI Consulting — Uskudar

Uskudar leans on the health, education and service institutions around Altunizade and Baglarbasi. If you keep your records in an archive of your own, the AI question starts there: how does accumulated material become searchable? The work closest to this district's sector profile is Vipkurum's question cropping tool. It was written for exam content. We have nothing that parses a patient file, an application form or institutional correspondence, and we make no promise in that area. What carries across is how the pool is built: the page is broken up and each piece is written to its own record. Cropped images and question records live in MongoDB, and the Python service reaches Laravel's content side over Node.js.

  • A booklet page is divided into single questions and each one sits in the pool on its own record, so material that was stored whole becomes searchable.
  • When the pool has no match, the questions already on hand serve as examples: the language model writes a new one in the same form and produces its answer with it.
  • We keep the scope visible: parsing patient files, application forms or institutional correspondence is not work we have done.

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

Vipkurum question cropping — a whole booklet becomes a searchable question pool

Vipkurum's question cropping tool turns a booklet stored whole in an archive into a usable pool. OpenCV divides the page into single questions, each is written to its own record, and Gemini and OpenAI attach the answer from the key. Where the pool has no match, existing questions act as examples: the model writes a new one in the same form and produces its answer with it. A fresh worksheet then draws on material already held, and no page goes to the copier. Cropped images and question records live in MongoDB, and the Python service reaches Laravel's content side over Node.js. The context is exam content, not institutional records.

Read the case study →

Working with Uskudar 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

  • Our archived booklets are scanned images. Is that usable as it stands?

    An image is what the tool expects. The page image is divided into single questions and each piece is written to its own record, so nothing needs to exist as an editable document first.

  • Do you hand our institutional data to a model?

    What reaches the model is a question and an answer key. The record side stays in MongoDB, and an institution's own data is not part of this flow. Nor have we built anything that parses patient files or application forms.

  • What happens when we need a question the pool does not hold?

    The questions already held are used as examples and the model writes a new one in the same form, producing its answer with it. The result is not a file appended afterwards; it arrives with its own record.

  • Would you not need to visit the institution and see the archive?

    We keep no office and work fully remotely. Rather than look through an archive, we first write down which fields a question is stored with, because the pool is searchable only to the extent those fields are decided.

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