# Aki Gogikar

> Akhilesh (Aki) Gogikar is the founder and Executive Vice President of Mendel Info Labs, where he leads product and engineering across OneNew and ActPass. He builds private, controlled AI systems around privacy-aware organizations and researches agent governance, organizational memory, efficient inference, machine learning systems and complexity.

Deploy at https://akigogikar.com — serve this file at https://akigogikar.com/llms.txt

## Names

- Professional name: Aki Gogikar — used across business, technical, and speaking contexts
- Author and legal byline: Akhilesh Gogikar — used on books, papers, and filings
- Both names refer to the same person; the canonical written form is "Akhilesh (Aki) Gogikar"

## Role

- Founder and Executive Vice President, Mendel Info Labs LLC (https://www.mendelinfolabs.com)
- Leads product and engineering across OneNew and ActPass

## What he builds

- [OneNew](https://onenew.ai): the single customer-facing offer for private AI built around each organization; its service model is designed to discover the requirement, design the controlled solution, connect approved data, provision a managed private-AI Workspace on customer-controlled infrastructure, and operate it with the customer, with a broader direction spanning Workspace, API, Desktop, organizational memory, private model access, and approved custom tools
- [ActPass](https://actpass.org): a separate, vendor-neutral governance product that evaluates consequential agent actions as allow, deny, or needs approval before execution

## Writing

- Author, *Linearity, Non-linearity and Chaos* (byline: Akhilesh Gogikar)
- Author, [Adaptive Runge-Kutta Step Control Buys Training Loss, Not Generalization](https://arxiv.org/abs/2607.14516), a compute-matched RK-Adam optimizer study
- [RK4Optimizer](https://github.com/akigogikar/RK4Optimizer): code, data and figures for the RK-Adam study
- [Medium archive](https://medium.com/@akhileshgogikar): essays on artificial intelligence, data and systems

## Topics for interviews and collaboration

- Private AI services built around organizations
- Deterministic permission boundaries for AI agents
- Customer-controlled infrastructure and organizational memory
- Efficient private inference
- Reproducible machine learning research

## Elsewhere

- LinkedIn: https://www.linkedin.com/in/akigogikar
- GitHub: https://github.com/Akhilesh-Gogikar
- Hugging Face: https://huggingface.co/AkiGogikar
- Medium: https://medium.com/@akhileshgogikar
- arXiv: https://arxiv.org/abs/2607.14516

## Contact

- [Email](mailto:aki@onenew.ai): pilots, press, podcasts, collaboration