Jyotsna G profile photo 1 on Tal

Jyotsna G

Building Kai — Enterprise Agentic OS | IIT Madras’26

Bengaluru

1+ year of work experience

Kai Trust
Sep 20262 mosPresent

Building Kai, Agentic Operating System for enterprises. Kai helps organisations move from scattered AI experiments to governed, enterprise-wide adoption. It builds an organisation brain, connects to internal systems, and lets teams chat with their knowledge, build internal apps, automate workflows and run business playbooks, with the visibility and control large companies need to scale AI responsibly. I work across product and go-to-market initiatives such as: • Shaping Kai's business playbooks: turning recurring processes in HR, legal, finance, operations and GRC into agentic workflows that stay observable, governed and human-aware. • Building a community of AI-first leaders, starting with HR, where members turn their team's processes into playbooks, reuse what peers have built, and share their own. • Talking with CHROs, CAIOs, CIOs, legal and operations leaders about where AI adoption actually stalls, and turning that into concrete plays across discovery, build, enablement and scale. If you lead AI adoption, HR, legal, sales or operations in a large organisation and are figuring out how to move from pilots to real workflows, I'd love to compare notes. My DMs are open.
Sprinto
Jul 20264 mosPresent

At Sprinto, I contributed to the development of Sprinto Assist, a suite of browser agents that automate compliance tasks. I focused on improving the accuracy of the Fix-It agent and enhancing the Security Questionnaire agent's answer quality. My efforts led to significant reductions in false reports and streamlined issue resolution across multiple infrastructure checks.
Sprinto
Apr 20263 mosJun 2026

Enhanced Sprinto’s Autonomous Trust Platform by improving the agentic experience, architecting and shipping production-grade features that streamlined compliance operations.
Indian Institute of Technology, Madras
Dec 20258 mosJul 2026

Chennai

Application No.: 202641098611 (IDF 3873), titled “SYSTEM AND METHOD FOR AUTOMATED REPORT GENERATION FROM MULTIMODAL INPUT DATA”, published on the Indian Patent Office (IPO) website, in Official Journal No. 34/2026. Backed by the Nirmaan pre-incubation cell, IIT Madras.
The University of Texas at Austin
May 20254 mosAug 2025

Austin, Texas, United States

• Conducted extensive research on model-free methods for control implementation in robotic systems, reviewing over 30 academic papers. • Achieved autonomous control of a 5-DOF robotic arm using SAC and PPO algorithms for 3D trajectory determination. • Attained a minimal tracking RMSE error of 0.0035 through the application of Model Predictive Control (MPC) and physics-driven modeling.
Siemens
Jan 20255 mosMay 2025

Bengaluru, Karnataka, India

• Co-authored a whitepaper on Agentic-AI, focusing on agent interoperability and market insights about MCP and A2A protocols. • Enhanced risk classification accuracy from 68% to 84% through the development of Agentic-RAG for Legal Operations Assistant in Risk classification. • Achieved 91% accuracy in the Intellectual Property Chatbot by implementing a self-reflection mechanism.

Jyotsna’s AI Stack

claude code
claude code
claude
claude
CrewAI
CrewAI
Fireflies.ai
Fireflies.ai
Napkin AI
Napkin AI
Wispr Flow
Wispr Flow

Education

Indian Institute of Technology, Madras

Dual degree (Btech+Mtech), Engineering Design

Indian Institute of Technology, Madras

20215 yrs2026

Join Jyotsna and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Jyotsna G on Tal

Headline

Building Kai — Enterprise Agentic OS | IIT Madras’26

About

I graduated from IIT Madras in July 2026 and have spent the rest of this year building ai features in the compliance space at Sprinto. Now, at Kai, I build and design agentic playbooks that automate workflows for enterprises. Bringing AI into how your teams work? I'd love to hear what you're trying to automate. Say hi or visit kaitrust.com

Location

Bengaluru

AI stack

claude code

claude

CrewAI

Fireflies.ai

Napkin AI

Wispr Flow

Work experience

Founder's office at Kai Trust, Sep 2026 - Present · 2 mos, Building Kai, Agentic Operating System for enterprises. Kai helps organisations move from scattered AI experiments to governed, enterprise-wide adoption. It builds an organisation brain, connects to internal systems, and lets teams chat with their knowledge, build internal apps, automate workflows and run business playbooks, with the visibility and control large companies need to scale AI responsibly. I work across product and go-to-market initiatives such as: • Shaping Kai's business playbooks: turning recurring processes in HR, legal, finance, operations and GRC into agentic workflows that stay observable, governed and human-aware. • Building a community of AI-first leaders, starting with HR, where members turn their team's processes into playbooks, reuse what peers have built, and share their own. • Talking with CHROs, CAIOs, CIOs, legal and operations leaders about where AI adoption actually stalls, and turning that into concrete plays across discovery, build, enablement and scale. If you lead AI adoption, HR, legal, sales or operations in a large organisation and are figuring out how to move from pilots to real workflows, I'd love to compare notes. My DMs are open.

Associate Product Manager at Sprinto, Jul 2026 - Present · 4 mos, At Sprinto, I contributed to the development of Sprinto Assist, a suite of browser agents that automate compliance tasks. I focused on improving the accuracy of the Fix-It agent and enhancing the Security Questionnaire agent's answer quality. My efforts led to significant reductions in false reports and streamlined issue resolution across multiple infrastructure checks.

Technical Product Manager at Sprinto, Apr 2026 - Jun 2026 · 3 mos, Enhanced Sprinto’s Autonomous Trust Platform by improving the agentic experience, architecting and shipping production-grade features that streamlined compliance operations.

Founder Medyx.ai at Indian Institute of Technology, Madras, Dec 2025 - Jul 2026 · 8 mos, Chennai, Application No.: 202641098611 (IDF 3873), titled “SYSTEM AND METHOD FOR AUTOMATED REPORT GENERATION FROM MULTIMODAL INPUT DATA”, published on the Indian Patent Office (IPO) website, in Official Journal No. 34/2026. Backed by the Nirmaan pre-incubation cell, IIT Madras.

Research Associate at The University of Texas at Austin, May 2025 - Aug 2025 · 4 mos, Austin, Texas, United States, • Conducted extensive research on model-free methods for control implementation in robotic systems, reviewing over 30 academic papers. • Achieved autonomous control of a 5-DOF robotic arm using SAC and PPO algorithms for 3D trajectory determination. • Attained a minimal tracking RMSE error of 0.0035 through the application of Model Predictive Control (MPC) and physics-driven modeling.

Data and Artificial Intelligence Intern at Siemens, Jan 2025 - May 2025 · 5 mos, Bengaluru, Karnataka, India, • Co-authored a whitepaper on Agentic-AI, focusing on agent interoperability and market insights about MCP and A2A protocols. • Enhanced risk classification accuracy from 68% to 84% through the development of Agentic-RAG for Legal Operations Assistant in Risk classification. • Achieved 91% accuracy in the Intellectual Property Chatbot by implementing a self-reflection mechanism.

Education

Dual degree (Btech+Mtech), Engineering Design, Indian Institute of Technology, Madras