Gargi Kshirsagar

From Enterprise Data to AI Agents: Building Practical AI Solutions with Open Source Models

Submitted Jun 23, 2026

Description -
Generative AI is rapidly moving from experimentation to real-world enterprise adoption, but organizations still face challenges around data privacy, cost, model dependency, and integrating AI into existing workflows. This session explores the practical journey of building AI-powered solutions using open-source models, focusing on how open-weight models can be combined with enterprise data platforms and applications to create useful AI assistants and automation workflows.

Takeaways -
The session will share hands-on learnings from implementing AI solutions in enterprise environments — including connecting models with internal data, improving responses through domain knowledge, evaluating model performance, and understanding the engineering challenges behind production AI. Attendees will gain practical insights into what works, what does not, and how open-source AI can become a viable option for building scalable, secure, and cost-effective AI solutions.
Understand practical approaches for using open-source AI models with enterprise data to build AI assistants and automation workflows.
Learn key lessons from taking AI solutions from experimentation to production, including model selection, integration challenges, performance trade-offs, and real-world implementation considerations.

Audiences -
This session will be beneficial for software engineers, data engineers, data scientists, AI practitioners, and technology leaders who are exploring how to apply open-source AI models in real-world enterprise scenarios. It is especially relevant for professionals interested in building AI agents, integrating LLMs with business systems, improving productivity through automation, and understanding the practical considerations of deploying AI solutions in production environments.

Speakers
Gargi Kshirsagar — Software Developer, IBM (gkshirsa@ibm.com)
Software Developer at IBM, working on enterprise data, AI solutions, and automation initiatives. My work involves building practical AI-driven solutions, including AI agents that integrate with enterprise systems and data platforms to improve business workflows and user productivity.

Hrithik Gavankar — Software Engineer, Red Hat (hgavanka@redhat.com)
Software Engineer at Red Hat (Ansible), building web systems, developer tools, AI agents, and MCP (Model Context Protocol) systems. Passionate about open-source technologies and creating scalable solutions that improve developer productivity and automation workflows.

Presentation link - https://docs.google.com/presentation/d/1KCZpVPX0dCTlSS-Ics9Kz7B4sZmFwkd6/edit?usp=drivesdk&ouid=100257767408290791874&rtpof=true&sd=true

Comments

{{ gettext('Login to leave a comment') }}

{{ gettext('Post a comment…') }}
{{ gettext('New comment') }}
{{ formTitle }}

{{ errorMsg }}

{{ gettext('No comments posted yet') }}

Hosted by

Jumpstart better data engineering and AI futures