LLMOps in Production: Observability, Evals & Guardrails

Moving Large Language Models from prototype to production isn't just about prompt engineering—it’s about Observability, Evals, Safety, Alignment, and Governance. Join the Future of Data Bangalore developer community for an in-person, technical deep-dive into building production-grade LLM applications. Host & Community Partners Cloudera, Vivanti, and Globalsinc are bringing together AI engineers, ML platform architects, and open-source contributors across Bangalore to share real-world engineering strategies. ## 📍 Event Details Date: Saturday, 21st November 2026 Time: 9:30 AM – 2:00 PM IST Host: Cloudera Partners: Cloudera + Vivanti + Globalsinc ## 🗓️ Agenda & Schedule 9:30AM to 10:00AM: Registration & Welcome Coffee 10:00AM to 10:30AM: Keynote & Community Intro 10:30AM to 11:00AM: Talk 1 TBD 11:00AM to 11:30AM: Talk 2 TBD 11:30AM to 12:00PM: Talk 3 TBD 12:00PM to 12:30PM: Panel Discussion: Real-world LLMOps pitfalls & Successes 12:30PM to 12:45PM: Q&A 01:00PM to 2:00PM: Lunch & Networking ## 🎯 Core Discussion Topics Whether you are fighting prompt drift, setting up eval pipelines, or implementing guardrails before your next release, our sessions focus on four core pillars: 1. Observability & Tracing: Tracking prompt trajectories, token costs, latency, and context degradation across multi-step chains. 2. Evals & Benchmarking: Automated scoring pipelines, synthetic test datasets, and continuous evaluations in production. 3. Guardrails & Safety: Enforcing real-time input/output sanitization, schema validation, and PII detection. 4. Governance & Alignment: Auditing model behaviors, maintaining compliance, and managing model versions across environments. ## 🤝 Community & Support A massive shoutout to Cloudera for hosting us and providing lunch, as well as Vivanti and Globalsinc for partnering with us to support the local developer ecosystem. ⚠️ Important Note: In-person seats are limited to ensure a quality networking experience. Please RSVP early to secure