Dhivya Nagasubramanian leads AI transformation at a major U.S. financial institution. She’s not just a corporate exec—she’s a patent holder in applied machine learning and author of Agentic AI for Engineers. Her book has surpassed 6,000 institutional accesses on SpringerLink and sits in over 260 libraries worldwide.
Her research digs into building AI applications that resist adversarial jailbreak attacks. She also works on developing models for multicultural safety and security. Nagasubramanian warns about risks in regulated finance: “The most dangerous failure is a wrong number that looks right.” She insists true automation means mapping every possible system path before running it.
At Uber, CTO Praveen Neppalli Naga has deployed “Agentic Pods”—teams of top AI engineers embedded within departments like finance, legal, marketing, HR, and procurement. These engineers spend days shadowing staff before designing AI tools. The results are dramatic: financial planning time dropped from 15 hours to 30 minutes. Generating financial reports shrank from two days to 10 minutes. Marketing quality checks went from two weeks to under an hour.
Tech executives compare this to Silicon Valley’s “forward-deployed engineer” role. Vantortech’s chief product officer, Peter Wilczynski, calls it the “Rearward Deployed Engineer.” Uber is not alone in embedding AI experts directly into business units to speed innovation and fix longstanding inefficiencies.
BusinessNext COO Pulkit Midhra highlights a partnership with ServiceNow that builds an autonomous “front-to-back” banking layer. This alliance aims to automate entire banking workflows seamlessly—no human bottlenecks allowed.
Meanwhile, Anthropic’s product executive Dianne Penn uses AI to sharpen management skills. She created a custom Claude “skill” inspired by the book Crucial Conversations. Penn uses Claude to prepare for difficult workplace talks and brainstorm appropriate responses. “What you don’t want is an AI that just agrees with you,” she says. “You want this technology to augment and grow to a better outcome. Sometimes having Claude push back makes me better.”
Penn shares this approach with other managers at Anthropic, encouraging teams to use AI for better conversations and stronger coaching. She stresses Claude is a thinking partner, not a crutch. The AI refines ideas, it doesn’t replace human thought.
On a different front, Sanghita Dey, founder and CEO of SolusFinance, focuses on the human side of trading. With over 15 years in tech, financial services, and AI, Dey sees most traders fail despite access to data and tools. The problem, she explains, is the gap between information and action.
SolusFinance combines trading competitions, AI coaching, and behavioral analytics to close that gap. Dey has filed intellectual property on behavioral intelligence generation and adaptive coaching. Her philosophy: technology should empower people, not just automate them. She predicts AI will become a powerful partner in financial decision-making, not a replacement.
These developments show AI shifting from isolated experiments to embedded, people-focused innovations. Whether cutting finance reporting from days to minutes or helping managers hold tough conversations, AI’s role is evolving. It’s not just automation. It’s transformation—with a human face.
Based on
- Dhivya Nagasubramanian, VP of AI Transformation and Innovation – Interview Series — unite.ai
- After starting the tokenmaxxing panic, Uber’s CTO is back with a very different AI story | Business Insider Africa — africa.businessinsider.com
- Inside Sanghita Dey’s quest to optimize the human element of trading | VentureBeat — venturebeat.com
- BusinessNext COO on ‘win-win’ AI banking alliance with ServiceNow — cnbc.com
- Anthropic exec shares how she uses AI to help manage her team | Business Insider Africa — africa.businessinsider.com
