TestMu Conference 2026 Kicks Off, Focusing on Agentic AI

Colantonio has hosted every edition of the event since its inception.

He highlighted the significant growth the conference has experienced over the years, underscoring its increasing relevance in the quality engineering landscape.

This name change reflects a commitment to the evolving field of artificial intelligence within quality assurance.

Attendees can expect numerous speakers representing TestMu AI throughout the event, reinforcing the new brand's presence and vision.

This substantial scale demonstrates the widespread interest in agentic engineering and quality.

The event incorporates various engagement opportunities, including prizes, networking sessions, and interactive features, with attendees encouraged to actively participate in the chat.

While 2025 was largely a period of experimentation and doubt regarding AI's practical uses, the current focus is on implementing and validating these technologies. The event aims to equip testers with the skills needed to manage agentic workflows effectively.

Key technical areas covered at the conference include how to test AI, perform evaluations, and engage in prompt engineering. Participants will learn how to validate agentic builds and seamlessly integrate them into existing quality assurance workflows.

The curriculum also addresses the critical need for building trust in AI systems by emphasizing the use of guardrails and proper evaluation techniques.

Testers are encouraged to embrace AI and become leaders within their organizations by guiding management on its capabilities and limitations. Their role is crucial in demonstrating how to properly test AI to ensure effective integration.

Proper testing is essential to prevent adverse customer and business impacts, making testers vital in validating AI solutions. The overarching goal is to effectively communicate AI's potential and practical application to management.

The conference emphasizes that AI in quality engineering has transcended mere hype and is now delivering on its real-world promises. Joe Colantonio stated his observations from traveling worldwide, witnessing AI's practical implementation.

He highlights the importance of verifying solutions like Keen AI, which are demonstrating tangible results in practical testing scenarios.

The conference features a diverse lineup of industry experts from major companies like Microsoft and Replit, alongside leaders from Enterprise AI Advocacy. The speakers include both technical practitioners and high-level executives such as CEOs.

This diverse representation aims to provide a holistic view of AI's capabilities and limitations in testing, offering insights from various organizational perspectives.

Andrew Knight, widely known as Automation Panda, is a featured speaker recognized for his honest and non-hyped approach to AI implementation. Colantonio recently interviewed Knight about his practical use of AI within his development team.

Knight's presence underscores the conference's commitment to showcasing real-world applications of AI from those who implement it in their daily work.

The selection of appropriate AI models and effective prompt engineering are critical factors that impact both cost and performance. Joe Colantonio emphasized that choosing the right model for specific activities significantly affects operational expenses.

Different models serve varied purposes; for instance, low-level models may be better suited for test generation, while front-end models could be optimal for planning. Consequently, managing token prices is becoming a central concern for engineers in this evolving landscape.

This integrated approach adds significant value by alleviating the burden on users.

Cost-cutting strategies also include caching results to avoid redundant model runs, reducing unnecessary computations and associated expenses. Additionally, leveraging cost-efficient models such as Gemini Flash for specific tasks offers another avenue for optimizing resource allocation and minimizing expenditure.

The conference offers over 80 free sessions across three days, covering diverse topics such as agentic workflows and autonomous quality engineering. This extensive program provides ample opportunities for professional growth.

Attendees are encouraged to explore sessions beyond their immediate expertise, as even a single actionable takeaway can significantly accelerate their careers or improve business techniques.

Optimizing LLM costs hinges on strategically selecting the appropriate model for each specific activity. Upcoming sessions will delve into various aspects of LLM cost optimization.

Panels will feature discussions on quality engineering applications across diverse industries, including retail, finance, and healthcare, highlighting tailored solutions for sector-specific challenges.

Evaluation should therefore depend heavily on specific organizational needs.

A recommended approach involves conducting a two-week proof of concept sprint to validate a tool's suitability within a team's distinct operational context. This method ensures that chosen solutions are tailored and effective.

Joe Colantonio suggests that AI may automate the validation of code without extensive manual input, replacing repetitive tasks like abstracting page objects.

While current AI capabilities often function more as 'sanity checks,' they free up human testers to concentrate on business domain knowledge and critical thinking. This transition empowers testing experts to leverage AI for mundane automation while dedicating their expertise to high-value, oversight functions.

The upcoming keynote, titled 'The Era of AI: From QA to Trust Engineering,' is a highly anticipated session that will delve into the critical evolution of artificial intelligence. Joe Colantonio expressed excitement about this address, encouraging attendees to stay for its insights.

Participants are also urged to visit exhibition booths and attend raffles, fostering opportunities for networking with like-minded practitioners and engaging further with the conference offerings.