Human-in-the-Loop AI: Building Systems That Align with Privacy, Ethics, and Regulation

#BuildingWithAI Human-in-the-Loop AI Systems That Align with Privacy, Ethics, and Regulation

What's This Meetup All About?

On Saturday, August 2, 2025, Covasant Technologies hosted the fourth edition of its much-anticipated #BuildingWithAI Meetup, themed "Human-in-the-Loop AI: Building Systems That Align with Privacy, Ethics, and Regulation." This event brought together enterprise leaders, AI practitioners, and data experts to discuss the evolving expectations from AI systems and the growing need for human oversight, ethical safeguards, and strong data foundations.

Attendees arrived and connected over introductions, sharing their interests in areas ranging from generative AI applications to LLM monitoring and AI governance. The energy in the room set the tone for a deeply engaging and forward-looking afternoon.

Senior Director of Covasant Technologies, opened the sessions with a compelling talk titled "What Enterprises Really Expect from AI Teams Today." Drawing from real client conversations, Covasant highlighted the shifting enterprise mindset from experimentation to accountable delivery. He emphasized the importance of explainability, outcome alignment, and responsible scaling as table stakes for enterprise AI success.

Next, Sitaram Tadepalli, SVP, Machine Learning Engineering at DBS Bank India, delivered an insightful session on "Privacy, Ethics, and Governance in the Age of Agentic AI." He explored how privacy-aware architecture, policy-as-code enforcement, and regulatory compliance are being embedded into real-world AI workflows. His talk underscored the importance of structured governance, especially as agentic AI systems gain autonomy.

After a short networking break, CV Rammohan, Chief Delivery Officer at Covasant Technologies, led the session "Prompts, Probabilities & Prejudices – Why AI Still Needs the Human Touch." He walked attendees through real-world examples of AI hallucinations, biased outcomes, and false positives emphasizing why human context remains critical even in the most advanced generative models.

The final session was delivered by Subhash Konduru, Senior Technology Manager at a leading UK-based multinational bank. In his talk, "Data Foundation for AI – Building Enterprise Identity for Hyper-Personalization," Subhash shared strategies for engineering consent-aware data pipelines and constructing identity graphs that drive personalization while upholding privacy and compliance.

The meetup concluded with a vote of thanks and closing remarks. Participants left with actionable insights, new connections, and a shared vision: to build AI systems that are not just powerful, but also trustworthy, transparent, and human-aligned.