fpga
(6)
Eriko Nurvitadhi Is Fixing the Expensive Pause Between GPUs
The AI boom bought more GPUs. Eriko Nurvitadhi is working on the less glamorous problem: keeping those expensive machines fed, connected and busy.
Gilad Tal
Gilad Tal is co-founder and CEO of DualBird, a Boston-based deep tech startup rebuilding the foundation of analytics and AI data infrastructure by running silicon acceleration as cloud-native software. A Technion-trained electrical engineer and physicist with 14 years in ASIC and software development, he previously founded and led TerrainEDA, an EDA startup acquired by Synopsys. At DualBird he and his co-founders use rewritable FPGA hardware to plug into Apache Spark and Iceberg environments, claiming up to 100x faster performance and up to 90% lower cost with no changes to existing workflows.
Khachik Sahakyan
Khachik Sahakyan is the co-founder and CEO of Grovf, a deep-tech hardware company building AI back-end fabric IP, chiplets and 400/800 Gbps SmartNICs that move data between AI accelerators with sub-2 microsecond RDMA latency. A microwave engineer and former Yerevan State University lecturer turned founder, he splits Grovf between Newark, California and Yerevan, Armenia, and has spent the last several years convincing data-center buyers that the path to scale is in silicon, not software.
Debbie Marr
Dr. Debbie Marr is the co-founder, CEO and President of AheadComputing, a Beaverton, Oregon startup building high-performance 64-bit RISC-V CPU cores. A former Intel Fellow and Chief Architect of Intel's Advanced Architecture Development Group, she spent 33 years shaping processors from the i386 era to modern Core and Xeon chips, served as server architect of the Pentium Pro, turned Hyper-Threading from concept into shipping product on the Pentium 4, and was chief architect of the 4th Gen Core (Haswell). She holds a PhD from the University of Michigan, more than 40 patents and 30+ published papers, and in 2024 left Intel with three fellow architects to prove that open architectures can deliver top-tier performance because, in her words, everybody deserves better compute.
Frank Ferrante
Frank Ferrante is the CEO of ForwardEdge ASIC, a Lockheed Martin subsidiary dedicated to advancing U.S. domestic semiconductor chip design. A 30-year veteran of the semiconductor industry, Ferrante has held leadership roles at Intel (Senior Director of Military, Aerospace and Government), Wolfspeed (VP of Worldwide Automotive Sales and Marketing), and Altera. He is a vocal advocate for U.S. advanced manufacturing policy, having advised policymakers on the CHIPS Act, SHIP, and RAMP-C initiatives. Under his leadership, ForwardEdge ASIC achieved U.S. Government Trusted IC Vendor Status in February 2026 and selected the MIPS S8200 for mission-critical autonomous platform ASICs.
Peter Morales
Peter Morales is the CEO and co-founder of Code Metal, a Boston-based AI company that hit unicorn status in February 2026 after raising a $125M Series B led by Salesforce Ventures. With a background spanning BAE Systems (F-35 ML optimization), MIT Lincoln Laboratory (counter-drone AI), and Microsoft (HoloLens computer vision), Morales built Code Metal to solve a gap he discovered firsthand: AI can generate code fast, but mission-critical industries — defense, automotive, semiconductors — need proof that code is correct before it ships. Code Metal's verifiable AI platform translates high-level code into hardware-optimized, mathematically verified production code for edge devices, drones, FPGAs, and autonomous systems. Customers include the U.S. Air Force, L3Harris, Raytheon, Toshiba, and Bosch.