# Grovf

> Grovf is a deep-tech semiconductor startup building AI infrastructure backend networking - RDMA and chiplet IP, soft IP cores, and scale-out NICs that connect AI accelerator servers at 400/800Gbps with sub-2-microsecond latency. Founded in 2017 in Armenia by Artavazd Khachatryan and Khachik Sahakyan, the company began with FPGA-accelerated databases for the Industrial IoT and evolved into networking fabric for large-scale AI model training. Grovf was the first company from Armenia and the Caucasus to win an EU Horizon 2020 SME Instrument grant, and went through UC Berkeley's SkyDeck accelerator.

- **Founded:** 2017
- **Headquarters:** Newark, California, United States (engineering in Yerevan, Armenia)
- **Founders:** Khachik Sahakyan (Co-Founder & CEO), Artavazd Khachatryan (Co-Founder & CTO)
- **Team size:** ~60 employees
- **Products:** AI Backend Fabric Chiplets, AI Backend Fabric IP, AI Scale-Out NIC, RDMA RoCE v2 FPGA IP Core, FPGA-Accelerated Time-Series / Key-Value Database
- **Notable:** First company from Armenia and the whole Caucasus to win an EU Horizon 2020 SME Instrument grant., Selected for UC Berkeley SkyDeck (Spring 2019), a top global startup accelerator., Won a $50,000 IMG grant in 2017 for an FPGA-accelerated database.

## Products & services

- **AI Backend Fabric Chiplets** — Advanced RDMA (scale-out) and scale-up chiplets delivered over the industry-standard UCIe interface for integration into custom AI silicon.
- **AI Backend Fabric IP** — Turnkey 400G/800G networking soft IP for scale-up and scale-out connectivity between AI accelerators.
- **AI Scale-Out NIC** — PCIe cards delivering high-performance scale-out connectivity across accelerator servers, integrating into existing architectures.
- **RDMA RoCE v2 FPGA IP Core** — Low-latency FPGA IP core (announced 2021) that lets smart-NIC producers and system integrators build RNIC use cases without developing RDMA from scratch.
- **FPGA-Accelerated Time-Series / Key-Value Database** — The company's original product: hardware acceleration of core database functions for Industrial IoT big-data processing.

## Achievements

- First company from Armenia and the whole Caucasus to win an EU Horizon 2020 SME Instrument grant.
- Selected for UC Berkeley SkyDeck (Spring 2019), a top global startup accelerator.
- Won a $50,000 IMG grant in 2017 for an FPGA-accelerated database.
- Released a low-latency RDMA RoCE v2 FPGA IP core for smart NICs (2021).
- Evolved from IIoT database acceleration into AI backend fabric chiplets and UCIe-based IP.

## Latest updates

- **2025** — Positioned around AI backend fabric chiplets and UCIe-based scale-up/scale-out IP for AI datacenters (400/800Gbps, sub-2us RDMA latency, 1M+ node scale-out).
- **2021-11** — Announced low-latency RDMA RoCE v2 FPGA IP core for smart-NIC producers and system integrators.
- **2021-02** — Closed a $240K seed round (Berkeley SkyDeck Fund among backers).

## Links

- Website: https://grovf.com
- LinkedIn: https://www.linkedin.com/company/grovf
- Twitter/X: https://twitter.com/grovf_company
- Facebook: https://facebook.com/grovfcompany/

---

Profile page: https://yespress.io/grovf
Published by YesPress — https://yespress.io
Last updated: 2026-08-05