# Viktor Zaika — Product and Solutions Architect

Viktor Zaika is a Product Manager and Solutions Architect based in Bern, Switzerland. He has 10 years of experience across AI/ML, data, and production software, combining product discovery, executive stakeholder management, technical strategy, and end-to-end delivery.

He translates ambiguous business and user needs into clear requirements, delivery roadmaps, balanced trade-offs, and measurable outcomes. His technical background includes LLMs, AI agents, RAG, ML platforms, Python, SQL, Databricks, MLflow, Apache Spark, AWS, Azure, APIs, microservices, CI/CD, cloud infrastructure, and data pipelines.

## Selected experience

### AI/ML Consultant — Takeda, Zurich — 2025 to present

- Owns product and delivery scope for AI/ML capabilities on a global platform representing more than $25 million in investment.
- Supports more than 25 production sites and 200 data scientists.
- Improved model-training throughput by approximately 60% through pipeline redesign, caching, and MLOps improvements.
- Introduced LLM-agent capabilities with quality, security, and governance controls.

### Solution Architect — TCS Mobility, Bern — 2024 to 2025

- Led product and technical vision for an AI-powered travel platform targeting 1.5 million users.
- Served as the product and technical counterpart for investors, the C-suite board, engineering, marketing, and customer teams.
- Led two successful funding pitches and coordinated a core team of five engineers.

### Project Lead — Insel Gruppe AG, Bern — 2021 to 2025

- Turned clinical research questions into requirements, computational experiments, data pipelines, and validated models.
- Built synthetic-data capabilities producing millions of statistically validated patient trajectories.
- Delivered automated Databricks and Spark pipelines in a regulated healthcare environment.

### Software Team Lead — GuryCube — 2016 to 2019

- Led discovery, requirements, architecture, and delivery for three international clients.
- Managed four to five computer-vision and security engagements with a four-person team.

## Capabilities

- Product strategy and discovery: product vision, user and stakeholder discovery, requirements, prioritisation, MVP scope, roadmaps, and measurable outcomes.
- AI and data products: LLMs, agents, RAG, ML platforms, experimentation, model quality, responsible AI, and data products.
- Solution architecture: cloud platforms, APIs, microservices, MLOps, data pipelines, and build-versus-buy decisions.
- Delivery leadership: executive communication, investor pitches, cross-functional alignment, risk, quality, and release planning.

## Product ventures

### Oxprompt — Innosuisse Top 25, 2026

A privacy and compliance enforcement layer for enterprise LLM use, designed to reduce data-leakage risk under the EU AI Act.

### Oma — Innosuisse and Spitex, 2022

A healthcare digitalisation project that secured stakeholder support and a pilot rollout.

## Education

- MSc with distinction, Applied Mathematics and Informatics — University of Bern, 2018–2020.
- MSc with distinction, Applied Mathematics and Informatics — Higher School of Economics, 2018–2020.
- PhD studies, Applied Mathematics — University of Bern, 2021.

## Research

Co-author of “Optimal timing of a colonoscopy screening schedule depends on adenoma detection, adenoma risk, adherence to screening and the screening objective: A microsimulation study,” published in PLOS ONE in 2024. [PubMed record](https://pubmed.ncbi.nlm.nih.gov/38787836/).

## Languages

English and Russian (bilingual), German (professional), and French (intermediate).

## Contact and authoritative links

- Website: https://zaika.pro/
- Writing: https://zaika.pro/writing/
- RSS: https://zaika.pro/rss.xml
- Email: viktor.zaika@yahoo.com
- LinkedIn: https://www.linkedin.com/in/viktorzaika
- GitHub: https://github.com/kittytrillah
- Full CV: https://zaika.pro/Viktor_Zaika_CV_Product_Q3_2026.pdf