The Story

Pixion was born out of desire for independence

Not just any kind, though. If it's to be meaningful, one needs to be good at what they do. That's why we put in the effort, and, consequently, only work with the extraordinary doing some extraordinary things.

We've successfully worked with a number of clients from different industries, from construction to retail, and built platforms for both consumer and internal use. If there's anything in common with all our clients, its our shared openness, trust and strive for excellence.

If you'd like to check out some of our best work, feel free to browse through some of our case studies.

Services & Expertise

Our wide and versatile skill set allows us to confidently tackle every task. We've proven ourselves in building digital products from scratch, redesigning and maintaining existing platforms, creating visual identities and constructing seamless user experiences - and we're only just getting started.

Technologies

Tools

Small Team, Big IdeasIf you're passionate about making an impact you’ve come to the right place.

How great work comes aboutBuilding success through open communication

We genuinely enjoy our work and approach every project meticulously and with enthusiastic curiosity. Our results have consistently been best achieved through open and direct communication, which is why we turned it into a foundation of how we do work. Resourcefulness, creativity and efficiency are only side-effects of a good setup.

Work

No buzzwords, just greatProduct Development

Keep ReadingCheck out our blogs

Vector Database Benchmark - Chroma vs Milvus vs PgVector vs RedisBenchmark the performance of Chroma, Milvus, PgVector, and Redis using VectorDBBench. This article explores key metrics such as recall, queries per second (QPS), and latency across different HNSW parameter configurations. The results highlight trade-offs in vector search performance. Insights from extensive testing provide a clearer understanding of database behavior, indexing efficiency, and real-world applicability.

by Luka Panic11 min read

Development

LLM Prompt OptimizationOptimize your LLM prompts for Retrieval-Augmented Generation (RAG) with the JSON-based, and few-shot techniques. The article explores strategies for reducing hallucinations, improving context alignment, and addressing formatting errors. Extensive testing demonstrates the benefits of structuring prompts with clear instructions, separating context chunks, and leveraging in-context learning examples.

by Luka Panic6 min read

Development