Listen to users, drive the product

Problem

Months of rote memorization of identical vocabulary, intrusive gamification, lack of learning adaptation, AI used as entertainment rather than an effective tool.

Solution

2. - Developed adaptive and scalable learning algorithm
3. - Created AI-based pipeline for corpus data collection and exercise generation
      - Enables unusual language pairs (Finnish-Croatian, for example)
      - Corpus data for standard pairs (Polish-English, for example) collected within days (prior to methodologist validation)
4. - Single-screen flow-based interface without excessive options or settings
   - Demo application tested with target audience
5. - A product vision, strategy, and roadmap have been developed
6. Current status: CEO attracts investments, CTO is working on a technological solution. I remain a part-time player in this startup.

Problem

Massive corporate product knowledge database. Complex DB structure, difficult to search for information. Clients often wait too long for answers. Need to quickly obtain relevant responses.

Solution

Knowledge base structured and expanded with AI from all written sources. AI-powered chatbot delivers instant answers. Accessible on desktop and mobile - field staff get info just as fast.

Problem

To launch a self-service platform for advertisers from scratch and redesign the internal platform for account managers and other employees of the company.

This should significantly affect the number of new offers being added, the effectiveness of current ones, payments, etc. and, as a result, revenue.

Solution

- The demo of the cabinet was successfully presented at the conference in Barcelona and aroused the interest of existing and potential customers
- Part of the new functionality has become a “killer feature” and a reason to start partnerships with advertisers who previously refused to cooperate
- Optimized processes between departments within the company

Problem

Company suffered major losses due to lack of basic monitoring and testing systems. Complex inter-department communication and non-optimized accounting required manual intervention for every client request, causing delays and revenue loss. RTB product managed by single person (CEO) with no dedicated development team. Advertiser platform in early stage, unable to support network's core functionality.

Solution

Problem

Low digital service penetration in logistics leads to suboptimal truck loading and route planning, resulting in long downtimes, low transportation efficiency, high cost per kilometer, and reduced carrier profitability. This slows the growth of the "Monopoly.Online" and causes profit decline in some areas.

Solution

Problem

The product developed chaotically. Due to the legacy of the code, it was impossible to quickly implement new features, expensive to maintain. Distracts the team from working on new functionality

Other There were too many manual steps in launching ads and integrating new partners. It was hard to scale without hiring more people, and managers couldn’t grow existing partners because they were overloaded with routine work.

Solution

- Successfully redesigned and rebuilt the SSP for publishers, resulting in faster feature delivery, reduced workload for support and finance teams, and increased ad revenue for website owners.

- Developed an in-house White Label user platform, reducing new product development time from 3-6 months to 1-3 months.

Other

- Saved 10-30% (depending on the partner network) of integration costs by introducing anti-fraud features and improving auctions in the RTB product.
- Launched a self-service platform for RTB account managers, resulting in cost savings of ~1M rubles/month and faster integration of new partners.