Piris Labs
Product & Competitive Intelligence
Builds photonic AI inference hardware, 5x lower latency and 10x lower power than GPUs.
Company Overview
Builds a vertically integrated AI inference platform combining proprietary photonic (optical) interconnects with a custom software stack to deliver ultra-fast, scalable, and cost-effective inference for trillion-parameter AI models.
Competitive Advantage & Moat
Product Roadmap & Public Announcements
Piris Labs has signaled plans for modular photonic interconnect hardware, a vertically optimized inference software stack, and partnerships with chip makers and ODMs to integrate optical data movement into existing data center architectures. Public positioning: 5x lower latency, 10x lower power per bit, and 2x lower cost per token versus GPU clusters. Working π Conversion Engine prototype (October 2025). SBIR government partnership secured. Early access waitlist stage.
Signals & Private Analysis
Conference appearances at photonics and AI hardware events suggest ongoing academic collaborations (MIT RLE, Stanford LINQS). Hiring patterns indicate a focus on photonics engineers and systems software developers. The advisory relationship with ex-Groq/NVIDIA President Mohsen Moazami signals potential enterprise go-to-market strategy development. Prof. Marc Baldo (MIT RLE Director) as advisor provides academic credibility. Stealth posture suggests they are protecting IP and timing a larger reveal around a working hardware demo.
Product Roadmap Priorities
Photonic interconnect-optimized inference engine that minimizes data movement bottlenecks to deliver ultra-low-latency token generation for trillion-parameter LLMs.
They built a special light-based data highway inside AI chips so massive AI models can think faster while using way less electricity.
It's like replacing a congested highway of delivery trucks (electrical wires) with a network of teleportation portals (light beams) so packages (data) arrive instantly without burning any gas.
Modular, horizontally scalable inference architecture that enables seamless scaling from single-node to multi-rack deployments without degradation in per-token latency or throughput.
They designed their AI system like stackable LEGO blocks so you can keep adding more power without everything slowing down or getting expensive.
It's like having a restaurant kitchen where adding more chefs actually makes every dish come out faster instead of causing everyone to bump into each other and slow down.
Hardware-software co-designed model optimization pipeline that automatically adapts large AI models for photonic inference, maximizing throughput and minimizing latency without manual model surgery.
They built a smart translator that automatically reshapes any AI brain to run perfectly on their light-speed hardware without anyone needing to manually tinker with it.
It's like a universal travel adapter that automatically reshapes itself to fit any country's power outlet so your devices just work wherever you go.
Company Overview
Key Team Members
- Ali Khalatpour, Co-Founder & CEO
- Keyvan Rezaei Moghadam, Co-Founder & President
Founder Force Multiplier
Ali Khalatpour holds multiple MIT degrees (BSc EE/Math, MASc, MSc Physics, PhD EE), is a Harvard- and Stanford-trained optical scientist who developed the first room-temperature terahertz semiconductor laser, and led the development of GUSTO's optical engine for NASA. 10+ years of directly relevant optical hardware R&D. Keyvan Rezaei Moghadam has a USC PhD in EE, 5 years at Meta as Technical Lead/Engineering Manager, and 2 years as Tech Lead at X (Twitter), with experience building 0-to-1 AI infrastructure. Advisors: Prof. Marc Baldo (MIT RLE Director) and Mohsen Moazami (ex-Groq/NVIDIA President).
Funding History
- 2025 | Ali Khalatpour and Keyvan Rezaei Moghadam co-found Piris Labs.
- 2025 | Working π Conversion Engine prototype completed.
- 2025 | SBIR government partnership secured.
- 2026 | Accepted into Y Combinator Winter 2026 batch.
Competitors
- Wafer-Scale / Custom ASIC Inference: Cerebras Systems, Groq, SambaNova Systems.
- GPU Cloud Inference: Together AI, Fireworks AI, Anyscale.
- Optical/Photonic AI Hardware: Lightmatter, Luminous Computing, Ayar Labs (optical interconnects), Celestial AI.
- Hyperscaler Inference: AWS Inferentia, Google TPU, Azure Maia.
