AI Safety & Reliability
The most dangerous AI isn't the one that fails — it's the one you can't tell is failing.
As intelligent systems move into production, the conversation is shifting from capability to control. How do you design agents that reason transparently, fail gracefully, and stay stable across complex workflows?
Join us for a conversation amongst builders working on systems that are careful by design and dependable in production. ⚙️
Panelists
Hailey Nguyen — AI Research @Worktrace AI
Arjun Subedi — CPO @Astha
Prof. PK Prasanna Kumar — Chief AI Trainer @Business Optima
Moderated by: Akash Saraf — CEO & Founder @TheAgentic
Sponsor
About TheAgentic:
TheAgentic provides complete infrastructure for production-ready AI agents, currently powering companies in biotech, healthtech, cybersecurity, and fintech where accuracy isn't optional.
TheAgentic’s Sanscritic is a universal reasoning layer for AI, separating how systems think from the models they run on. Instead of relying on fragile prompts or locked-in model behavior, Sanscritic lets you design, control, and own your reasoning logic across any model, workflow, or domain.
TheAgentic’s Launchpad program partners with early-stage startups to build AI products on our infrastructure at zero cost and zero equity.
TheAgentic’s Prometheus program lets domain experts and consultants turn their knowledge into revenue-generating AI products, without building or running a startup themselves
Host
About Sanscritic:
Sanscritic is building the reasoning layer for the next generation of AI systems. Through its NeoLab, Sanscritic develops meta-reasoning systems, domain foundation models, and infrastructure for more reliable, specialized, and verifiable AI. Join us to learn how Sanscritic is pushing beyond generic LLMs toward AI that can reason, evaluate, and operate in high-stakes domains.
Agenda
5:00 - 5:45pm: Registration and networking
5:45 - 6:00pm: Intro to TheAgentic and Sanscritic
6:00 - 7:00pm: Panel Discussion
7:00 - 8:30pm: Open networking