GEMESYS
GEMESYS GmbH is a Bochum-based deeptech hardware company developing a fully analog AI chip using memristor technology to enable on-device AI training and inference, targeting device manufacturers, sensor companies, and mobile OEMs for edge AI applications. Pre-revenue, founded 2023.
- Company typePrivate
- Founded2023
- HeadquartersBochum, Germany
- Headcount11–50
- GTM typeB2B
- OfferingHardware or Manufacturing
What GEMESYS does
GEMESYS GmbH is a Bochum, Germany-based deeptech hardware company founded in 2023 that is developing a fully analog AI chip inspired by the human brain's information processing mechanism. The company's core product uses memristor technology to enable both AI training and inference operations directly on edge devices, with company-reported claims of a 10,000x improvement in training efficiency over traditional digital AI chips (and 100-1,000x energy efficiency gains per a parallel source describing the EMULAITE sub-project). The chip's low power consumption, high data efficiency, and small form factor are designed to permit AI training on mobile devices and embedded sensors, eliminating dependence on cloud infrastructure.
GEMESYS operates in a pre-revenue R&D phase and has not yet commercialized its chip. The implied revenue model is hardware chip sales to device manufacturers, sensor companies, and mobile device OEMs for edge AI applications, supported by an enterprise field sales motion. The company has assembled a small team (11-50 employees) composed entirely of MS/PhD-level engineers and is co-led by four founders: Dr.-Ing. Dennis Michaelis (CEO), Moritz Schmidt (CFO), Dr.-Ing. Enver Solan (CSO), and Daniel Krüger (CTO). The company has secured non-dilutive grants from multiple German and European programs (EXIST, ESF, BMWK, EU, Grüne Gründungen.NRW) and approximately $9.1M in equity funding in November 2024 from a syndicate led by APEX Ventures and Atlantic Labs with participation from Sony Innovation Fund, Plug and Play, and NRW.BANK. A core research partnership with Ruhr University Bochum underpins the EMULAITE sub-project running through December 2026.
GEMESYS firmographics
Firmographics- Name
- GEMESYS
- Legal name
- GEMESYS GmbH
- Website
- https://gemesys.tech
- Company type
- Private
- Founded year
- 2023
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- GEMESYS GmbH is a Bochum-based deeptech hardware company developing a fully analog AI chip using memristor technology to enable on-device AI training and inference, targeting device manufacturers, sensor companies, and mobile OEMs for edge AI applications. Pre-revenue, founded 2023.
- Ownership category
- akta.pro rank
GEMESYS industry classification
Industry- Product category
- AI Semiconductor / Edge AI Chips
- NAICS
- Semiconductor and Related Device Manufacturing (334413)
- akta.pro primary industry
- Edge AI Hardware & NPUs (SoCs, modules, accelerators) (HDAAAJAC)
- akta.pro secondary industry
- Edge AI Deployment, Orchestration & OTA Model Updates (HDAAAJAI)
Keywords
Where GEMESYS is headquartered
LocationHeadquarters
- HQ city
- Bochum
- HQ country
- Germany
- HQ region
- Europe
Offices1 record
Markets served
GEMESYS business model
Business model- GTM type
- B2B
- Offering type
- Hardware or Manufacturing
- Cost components
- Technology or R&D, Personnel, Operations, Marketing or Sales, Infrastructure
Revenue model
- Hardware Chip Sales: Pre-revenue company developing analog AI chips. Revenue will likely be generated through sale of AI hardware chips to device manufacturers, sensor companies, and mobile device makers for edge AI applications.
Go-to-market motion1 record
Marketing channels3 records
GEMESYS product offering
Product offeringCore offering
GEMESYS designs and develops a fully analog AI chip inspired by the human brain, using memristor technology to perform both AI training and inference directly on edge devices. The product targets device manufacturers, sensor companies, and mobile device makers with claims of up to 10,000x training efficiency gains and 100-1,000x energy efficiency improvements versus digital AI hardware.
Product overview
GEMESYS is a hardware company offering a single unified product: an analog AI chip designed for edge AI applications. The chip is the world's first hardware capable of bringing both AI training and inference capabilities to the edge, inspired by the human brain's information processing mechanism. A notable R&D sub-project is EMULAITE (developed with Ruhr University Bochum), which demonstrates their approach to energy-efficient AI training using analog circuit emulation and memristors.
Differentiator
Problem solved
Functional benefit
Products and services
- GEMESYS Analog AI Chip A fully analog AI chip inspired by the human brain, designed to perform both AI training and inference on edge devices. The chip uses memristor technology and neural networks with high sparsity and low depth, and targets device manufacturers, sensor companies, and mobile device makers with claimed 10,000x training efficiency improvement vs. digital AI hardware.
- EMULAITE A revolutionary AI training approach using analog circuit emulation and memristors, developed in collaboration with Ruhr University Bochum, that replaces gradient-based learning to deliver 100-1,000x more energy-efficient AI training for sustainable edge AI.
Quantifiable outcome
- AI training efficiency increased by factor of 10,000
- +2 more outcomes
Companies that use GEMESYS
Customer profileIdeal customer profiles1 record
GEMESYS technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- No
- API docs
- API detail
Core technology
AI maturity
App detail
AI capability4 records
Feature5 records
GEMESYS partnerships and signals
Strategic signalPartnerships
One partnership is on record.
- Ruhr University Bochum (LPS - Chair for Production Systems)coreStrategic research partnership for the EMULAITE project developing revolutionary analog circuit emulation AI training technology. GEMESYS collaborates with the Chair for Production Systems at Ruhr University Bochum. Project duration: 01.09.2024 – 31.12.2026. Funded by Grüne Gründungen.NRW program.
Scale indicators5 records
Recent moves5 records
Expansion highlights5 records
GEMESYS competitors and assessment
Company assessmentDirect peers
- BrainChip Holdings: BrainChip develops the Akida neuromorphic processor for edge AI inference — directly comparable as a brain-inspired AI chip targeting low-power on-device intelligence, though without GEMESYS' training-at-edge claim.
- SynSense: SynSense builds fully analog neuromorphic processors and event-driven AI chips for always-on edge applications — the closest technology analog to GEMESYS' brain-inspired, memristor-based analog architecture.
- GrAI Matter Labs: GrAI Matter Labs (now GrAI) makes neurosparse processors for ultra-low-power edge AI inference — comparable as a brain-inspired, sparsity-focused AI chip for embedded devices.
- Rain AI: Rain AI develops analog in-memory compute chips for energy-efficient AI inference — directly comparable architecture approach (analog compute) targeting dramatic energy-efficiency gains versus digital accelerators.
- Mythic AI: Mythic builds analog compute-in-memory AI inference chips for edge devices — the most direct architectural analog to GEMESYS using analog matrix processing to slash energy per inference.
- Hailo: Hailo makes edge AI accelerators (Hailo-8, Hailo-15) for on-device inference in automotive, cameras, and industrial IoT — competing for the same edge AI silicon sockets GEMESYS targets, though using a digital architecture.
- Sima.ai: Sima.ai (formerly Syntiant) produces ultra-low-power neural decision processors for always-on edge AI — comparable target market (battery-powered smart sensors and edge devices) and similar power-efficiency thesis.
Broad incumbents
- Intel (Loihi / Hala Point): Intel's Loihi neuromorphic research chips and Hala Point system represent the incumbent R&D effort in brain-inspired computing — comparable technology direction, though not yet a commercial product at GEMESYS' target scale.
- IBM (TrueNorth / NorthPole): IBM's TrueNorth and NorthPole neuromorphic processors are foundational analog-inspired AI accelerators — comparable technology category, with IBM's research scale setting a long-term competitive benchmark.
- Qualcomm (AI Engine / Hexagon): Qualcomm integrates AI acceleration (Hexagon NPU) across mobile SoCs serving billions of devices — represents the entrenched digital incumbent that any edge AI chip (including GEMESYS') must displace or supplement.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat2 records
Key risks6 records
Key highlights7 records
Customer concentration
GEMESYS social profiles
Digital presenceGEMESYS financial estimates
Financial estimateRevenue estimate
Valuation estimate
GEMESYS leadership team
Management profileNumber of profiles
Profiles4 records
GEMESYS funding detail
Funding detailFunding overview
Funding rounds2 records
Investors7 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
GEMESYS M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about GEMESYS
What does GEMESYS do?
GEMESYS designs and develops a fully analog AI chip inspired by the human brain, using memristor technology to perform both AI training and inference directly on edge devices. The product targets device manufacturers, sensor companies, and mobile device makers with claims of up to 10,000x training efficiency gains and 100-1,000x energy efficiency improvements versus digital AI hardware.
Is GEMESYS a public or private company?
GEMESYS is a private company. It is classified as venture growth investor backed and is currently operating.
When was GEMESYS founded?
GEMESYS was founded in 2023. It employs 11 to 50 people.
Where is GEMESYS based?
GEMESYS is headquartered in Bochum, Germany, in the Europe region.
How does GEMESYS make money?
One revenue line is on record: hardware Chip Sales.
Who are GEMESYS's main competitors?
Direct peers on record are BrainChip Holdings, SynSense, GrAI Matter Labs, Rain AI, Mythic AI, Hailo and Sima.ai. Broad incumbents are Intel (Loihi / Hala Point), IBM (TrueNorth / NorthPole) and Qualcomm (AI Engine / Hexagon).
Does GEMESYS have an API?
No public API is recorded for GEMESYS.
What industry is GEMESYS in?
GEMESYS's product category is AI Semiconductor / Edge AI Chips. Its primary akta.pro industry code is HDAAAJAC, Edge AI Hardware & NPUs (SoCs, modules, accelerators), with a secondary code of HDAAAJAI, Edge AI Deployment, Orchestration & OTA Model Updates. Its NAICS code is 334413.