ET

About

Etched, founded in 2022, designs transformer-specific ASICs with a hard architectural bet: transformers are the dominant and durable abstraction for AI workloads, so the right move is to burn that assumption into silicon rather than preserve generality. Their first chip, Sohu, is a single-model ASIC built exclusively for transformer inference. The throughput numbers are significant - Etched claims over 500,000 tokens per second on Llama 70B and an order-of-magnitude improvement in both throughput and latency relative to NVIDIA's B200. The trade-off is explicit: Sohu cannot run non-transformer workloads, and the entire value proposition collapses if the architectural assumption does.

The performance claims, if they hold under production conditions, have direct implications for workloads where GPUs currently hit hard limits. Etched points to two in particular: real-time video generation models, where per-frame latency budgets are tight and sustained throughput requirements are high, and deep chain-of-thought reasoning agents, where long output sequences and large batch depths stress both memory bandwidth and end-to-end latency. Whether the claimed gains survive real deployment - across varied sequence lengths, batch sizes, quantization schemes, and serving topologies - is the evaluation question that matters most for operators considering adoption.

On the infrastructure side, Etched is partnering with Rambus on memory and interface technologies, which speaks to where the bandwidth and signaling bottlenecks sit in a transformer-optimized design. The company has raised $120 million and carries a stated valuation of $5 billion as of available reporting. Founders Gavin Uberti, Chris Zhu, and Robert Wachen lead the company out of the US.

Open roles at Etched

Explore 25 open positions at Etched and find your next opportunity.

ET

Electrical Engineer, Hardware Systems

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Inference Software Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Senior Layout PCB Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Front-End Power Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Firmware Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

ECAD/MCAD Symbol Librarian

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Emulation Software Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Physical Design Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Finance

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Manufacturing Test & Production Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Senior Firmware Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Front-End CAD Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Mechanical Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

PCB Hardware Validation Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Power Optimization Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

RTL Design Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Head of Legal

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Performance Modeling Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

Design Infrastructure Engineer

Etched

Cupertino, California, United States (On-site)

3w ago
ET

IC Package Engineer

Etched

Cupertino, California, United States (On-site)

3w ago

Similar companies

EL

ElevenLabs

ElevenLabs is an AI audio research and deployment company building voice AI systems that serve millions of developers, creators, and enterprises. The company's technical focus spans speech synthesis, voice cloning, multilingual voice models, and conversational AI agents. Their models support over 70 languages, with core capabilities in text-to-speech, sound effects generation, and voice agent deployment. The company is backed by Andreessen Horowitz, Sequoia, and other investors. The platform consists of three main products: ultra-realistic AI voices designed for clarity, expressiveness, and multilingual support; an Agents Platform that enables teams to deploy voice agents capable of listening, talking, and acting; and a Creative Platform focused on content localization, storytelling, and accessibility improvements. Primary technical domains include speech synthesis systems, voice cloning infrastructure, and conversational agent platforms built on Python and TypeScript. ElevenLabs serves businesses ranging from early-stage startups to large enterprises across multiple verticals: customer support, sales automation, education, video production, publishing, and accessibility applications. Named use cases include reading articles, voice-over generation, voice restoration for individuals with disabilities, and building intelligent agents for support, sales, and education workflows. The company's operational model emphasizes both research and production deployment, with infrastructure supporting content localization and audio-based applications at scale.

124 jobs
RA

Reflection AI

Reflection AI develops open foundation models targeting superintelligent autonomous systems, with current work focused on autonomous coding as a path to broader cognitive automation. The company combines reinforcement learning and large language models to build systems capable of handling most cognitive work on a computer, positioning autonomous code generation as the bottleneck to unlock that capability. The team includes contributors to AlphaGo, AlphaZero, PaLM, GPT-4, and Gemini, bringing production experience across game-playing RL systems and frontier language models. This background suggests familiarity with the trade-offs in training large-scale models - compute efficiency, sample complexity, and the operational challenges of running RL at scale alongside supervised pretraining. Reflection's stated objective centers on keeping superintelligence open and accessible through open foundation models. For inference practitioners, this implies potential work on model architectures, training infrastructure, and deployment systems designed for broad distribution rather than proprietary deployment. The autonomous coding focus suggests evaluation infrastructure for code generation, likely including metrics beyond pass@k - compilation rates, execution correctness, and performance characteristics of generated code under real-world constraints.

49 jobs