AI Takes Center Stage at JavaScript Conference Armenia 2025

December 1, 2025  12:25

The JavaScript Conference Armenia (JS Conf Armenia), held annually in Yerevan, has become one of the country’s most anticipated and influential events for software engineers and technology professionals. As one of the leading conferences of its kind in Armenia, it brings together hundreds of developers, engineering leaders, researchers, and industry experts to explore emerging technologies, exchange knowledge, and build meaningful professional connections.

Supported by major Armenian technology companies – including Krisp Technologies, Inc., Synergy International Systems, Ameriabank, and Mootq – the conference features technical presentations and Q&A panel discussions delivered by top experts in the field. Although open to the public, JS Conf Armenia primarily attracts professionals deeply involved in software engineering and related disciplines. Each year, several hundred participants attend to take advantage of the conference’s strong networking environment, designed to foster collaboration, promote innovation, and support professional growth. Over the years, JS Conf has firmly established itself as one of Armenia’s premier forums for skill development and industry engagement.

The 2025 edition of JS Conf Armenia, held on October 25, placed a strong spotlight on artificial intelligence (AI), reflecting both global technological trends and Armenia’s rapid advancements in the field. Many of this year’s speakers were leading experts in AI-driven technologies, offering attendees insight into the latest developments and evolving landscape of intelligent systems.

Below are three AI-focused presentations from the conference that are particularly noteworthy.

  • Armen Vardanyan “Building AI powered apps with NodeJS and Gemini”

Vardanyan introduced the audience to the concept of embeddings – numeric representations of text that position semantically similar meanings close to one another in vector space. This technique underpins semantic search, allowing systems to retrieve information based on meaning rather than exact keywords. He also demonstrated how embedding models can be combined with machine learning techniques to reduce computational costs while maintaining strong performance.

Без имени.jpg (350 KB)

Armen Vardanyan, a recurring speaker at JS Conf Armenia, is a Google Developer Expert for Angular, and a Team Lead at Volo – a multi-national software consulting company. He is also the author of the book Modern Angular, a comprehensive book covering new features and techniques of the most modern versions of Angular, and a panelist for the popular Adventures in Angular podcast.

  • Erik Davtyan“LLM Future is Multi-Model”

Davtyan’s talk explored recent research on multi-model AI, starting with automated model routing – techniques for selecting the optimal model for a given prompt. He then outlined advances in prompt optimization, showing how genetic learning algorithms can autonomously generate improved prompts that make AI systems more efficient and capable. His presentation signaled a future defined not by individual models but by dynamic AI ecosystems.

Без имени2.jpg (346 KB)

Erik Davtyan, also a returning speaker, is an applied AI expert. Most recently, he served as the only external judge at the Krisp Hackathon 2025. He currently works at Not Diamond, a San Francisco based startup developing intelligent infrastructure for multi-model systems. He has also been sought after by other distinguished U.S. tech startups, including A.Team and HearMe.

  • Rouben Meschian“Introduction to MCP”

Meschian presented a foundational overview of the Model Context Protocol (MCP), which is a protocol that allows AIs to use external tools. He walked attendees through manually implementing tool use, explained why a generalized protocol is essential for modern AI systems and demonstrated how to build and deploy MCP servers.

Без имени3.jpg (385 KB)

Rouben Meschian is one of the founding members of JS Conf Armenia. He is a software engineer with over 20 years of experience and a board member of YerevanN, one of Armenia’s leading AI research labs. He has also lectured at the American University of Armenia, Tumo Center for Creative Technologies, and taught in-depth programming courses on YouTube.

Collectively, these sessions underscored a central theme of the 2025 conference: AI is no longer an emerging trend but a transformative force actively reshaping how engineers design, build, and optimize software. As AI continues to evolve at an accelerated pace, it has become increasingly critical for technology and investment professionals to stay informed. Conferences like JS Conf Armenia highlight not only the rapid progress of the field but also the growing need to understand, navigate, and capitalize on the next wave of AI innovation.

Lilith Avagyan


 
 
 
 
  • Archive