We connect AI to the systems your business already runs, and build the new ones around it. Custom development and QA since 2009, for enterprises and technology companies.
building software since
Israel and Germany
web, mobile, desktop, embedded
from banking to agritech
Not a chatbot for its own sake. Each of these connects to your data, follows your access rules and is measured against the work it replaces.
Employees ask questions in plain language and get answers from contracts, policies and manuals, with links to sources.
Less searching, fewer questions to experts
Answers routine customer questions on your website, WhatsApp or Telegram and hands complex cases to people with context.
Round-the-clock support, same team size
Invoices, forms, applications and scans become structured records in your CRM, ERP or database, checked by a person.
Manual data entry replaced by review
Agents classify requests, draft replies, update your systems and route tickets, with people approving the key steps.
Routine work done faster and consistently
Search that understands meaning, not only keywords, across products, knowledge bases, tickets and documents for staff and customers.
The right answer, found first time
Test cases generated from requirements, self-healing UI tests and AI review of every release before it ships.
Wider coverage, shorter release cycles
Banks, insurers and clinics cannot afford experiments with their data. Every AI feature we build follows the same rules.
Models run in your cloud or on your servers when needed. Data is not used to train public models.
The assistant only sees what the user asking it is allowed to see.
Answers are checked against test sets before release and monitored after, so quality does not drift.
Decisions with real consequences are confirmed by an employee, and every action is logged.
Designed with GDPR and Israeli privacy requirements in mind from the first sketch.
We choose between OpenAI, Claude and open models per task and can switch when prices or quality change.
Language model features, assistants over your documents and agents for routine workflows.
Web, mobile, desktop and embedded applications built to your requirements.
Automated suites, AI-generated test cases and quality gates in every release.
Containers, pipelines and monitoring on AWS, Azure or your own servers.
Moving ageing systems to current platforms without stopping the business.
Architecture reviews, technology choices and a roadmap before you commit budget.
Projects for regulated, high-load businesses since 2009. We already speak your domain language.
AI document checks for onboarding, adviser portals and assistants that answer from regulations and procedures.
AI claims intake from photos and forms, policy search for agents, self-service apps for policyholders.
Customer portals and apps, AI ticket routing, support assistants and billing integrations.
Patient apps and appointment systems, AI search over medical records, built around strict privacy rules.
E-commerce, inventory and logistics, AI product search, recommendations and demand forecasts.
Professional AV, presentation devices and IoT: hardware, FPGA, firmware, software and AI on the device or in the cloud.
A few projects we can talk about. More under NDA on a call.
Presentation technology · WeKast
Task: turn a one-click screen-sharing device into a meeting tool that captures what was said and decided.
Solution: live transcription and AI summaries with action items sent to Slack or Teams, apps for four platforms and automated device testing.
from meeting end to summary
platforms from one team
Professional AV · PointyVision
Task: build one line of audio, video and control equipment for professional AV, broadcast and residential markets.
Solution: a shared architecture with modular routing on embedded Linux, FPGA design, firmware and control software, built by local and offshore teams.
video supported
layers, hardware to software
Agritech · Tevatronic
Task: turn soil sensor readings from thousands of hectares into irrigation decisions growers trust and act on.
Solution: machine learning that combines sensors with weather forecasts, a cloud dashboard and an AI assistant growers can ask in WhatsApp.
less water used
irrigation forecast ahead
Goals, users, data and where AI can save work, turned into scope.
You get: scope, estimate, risks
Architecture, interface prototypes and where AI fits in the process.
You get: prototype, architecture, test plan
Short iterations, AI-assisted code review, tests from the first build.
You get: regular demos, test reports
Release, monitoring and support once the software is live.
You get: live system, documentation, support
One use case, your real data, a working prototype and a clear answer on whether to scale it.
Agreed scope, timeline and budget. Best when requirements are clear.
Developers and testers working as part of your team, for long or evolving products.
An independent testing team for your releases, manual or automated.
Current, supported tools chosen for each project, not out of habit.
years in software development
projects delivered
engineers and testers
countries where our software runs
Tadiran, Clalit, WeKast, PointyVision, Tevatronic, Ticomsoft, Initech, RoyalApp
It stays in your infrastructure or in a cloud region you choose. For sensitive data we use models deployed in your environment. Data is never used to train public models.
Yes. We start with a code and architecture audit, then agree on what to fix first.
Yes. We work with clients in Israel and Europe, in English.
Book a free call. We will ask about the process you want to improve and suggest whether a pilot, a project or a consultation fits.
Yes, before we discuss any details of your project.
An engineer, not a sales script, replies within one business day.