What MakarenaLabs does

MakarenaLabs SRL is a privately held Italian ICT services and product company founded in December 2015 and headquartered in Verona, Italy, operating with 1–10 employees and a nominal share capital of €10,500. The company specializes in FPGA-based machine learning and digital signal processing, building solutions on AMD/Xilinx platforms including PYNQ, Kria SOM (KV260, KR260), Zynq Ultrascale+, and Zynq-7000. Its core product, MuseBox, is a real-time FPGA machine learning system offering pre-built AI tasks for face analysis, people analysis, object detection, text analysis, medical segmentation, and human segmentation, accessible via Python bindings and, since March 2023, a no-code browser GUI called CLIO. Supporting products include MRadio (audio stream analyzer), Mooseka (scalable audio analysis framework), and ALOE (augmented/virtual reality framework), alongside custom services in AR/VR, ML/DSP, hardware acceleration, demo development for trade shows, and teaching/training courses.

The company's revenue mechanics are mixed: bespoke ICT professional services engagements to enterprise clients in robotics, industrial automation, and healthcare/medical robotics are complemented by MuseBox product licensing on a node-locked evaluation basis, with self-service developer adoption via documentation and GitHub open-source repositories. GTM is sales-led and B2B, layered with a content-driven inbound strategy — extensive technical tutorials, an active GitHub presence (Smart IO, PM on SPYN, HLS examples, EDDP/SPYN, retro-console reference designs), a weekly newsletter, and occasional webinars. The company is a recognized AMD partner, has no disclosed external funding or parent ownership, and does not report revenue; its named management consists of CEO/CTO Enrico Giordano and CIO Matteo Castagnaro, both with the company since 2016.

Name

MakarenaLabs

Legal name

MakarenaLabs SRL

Website

https://makarenalabs.com

Company type

Private

Founded year

2015

Operating status

Operating

Headcount range

1–10 employees

Short description

MakarenaLabs is a small Verona, Italy-based ICT services and FPGA product company building real-time edge AI and digital signal processing solutions on AMD/Xilinx platforms, serving enterprise clients in robotics, industrial automation, and medical robotics through custom development and the MuseBox ML platform.

Ownership category

akta.pro rank

Product category

FPGA-based Edge AI Software

NAICS

Computer Systems Design and Related Services (54151), Navigational, Measuring, Electromedical, and Control Instruments Manufacturing (33451), Navigational, Measuring, Electromedical, and Control Instruments Manufacturing (3345)

SIC

Services-Computer Programming Services (7371), Services-Computer Integrated Systems Design (7373), Industrial Instruments For Measurement, Display, And Control (3823)

akta.pro primary industry

On-Device Inference Runtimes & SDKs (mobile/embedded) (HDAAAJAB)

akta.pro secondary industries

Edge AI Hardware & NPUs (SoCs, modules, accelerators) (HDAAAJAC), Real-Time Vision at the Edge (on-device CV/video) (HDAAAJAE), On-Device Speech & Audio AI (wake word, ASR, enhancement) (HDAAAJAF), AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ)

Keywords

• FPGA machine learning
• Hardware acceleration
• Edge AI inference
• Embedded signal processing
• FPGA development services

Headquarters

HQ city

Verona

HQ country

Italy

HQ region

Europe

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Marketing or Sales, Operations

Revenue model

1. Custom Software Development Services: ICT services including software development, AR/VR technologies, machine learning, hardware acceleration, and teaching/training courses. The company provides bespoke engineering services to enterprise clients.
2. MuseBox Product Licensing: Licensing of the MuseBox ML platform for edge AI applications. Evaluation licenses available for KV260 with node-locked licensing model.
3. Demo Development: Custom demo development for trade shows and product demonstrations.

Go-to-market motion1 record

Distribution channels2 records

Marketing channels5 records

Core offering

MakarenaLabs develops and licenses FPGA-based machine learning and digital signal processing software for edge AI applications on AMD/Xilinx platforms. Its flagship product, MuseBox, enables real-time audio and video AI inference (face analysis, human segmentation, object detection, medical segmentation, beamforming) through a no-code GUI (CLIO) and Python bindings. The company also delivers custom software development, hardware acceleration, AR/VR, and FPGA training services to enterprise clients in robotics, industrial automation, and healthcare.

Product overview

MakarenaLabs offers a portfolio of products centered on FPGA-based machine learning and signal processing. The flagship product is MuseBox, a real-time FPGA machine learning system that processes both audio and video through AI tasks (face analysis, human segmentation, animal detection). Supporting the audio AI stack are MRadio (audio stream analyzer) and Mooseka (scalable audio analysis framework). ALOE provides an artificial reality framework. All products leverage AMD/Xilinx FPGA platforms (Kria SOMs, ZYNQ) with DPU acceleration and are accessible via Python/Jupyter Notebook or a no-code GUI (MuseBox CLIO). The company also provides ICT services including software development, AR/VR, machine learning/DSP, hardware acceleration, and training.

Differentiator

Problem solved

Functional benefit

Brands

• MuseBox: Real-time FPGA Machine Learning system for AI pipelines, audio and video processing
• MRadio
• Mooseka
• ALOE

Products and services

• MuseBox Real-time FPGA machine learning system that runs AI tasks (face analysis, people analysis, object detection, text analysis, medical segmentation, human segmentation) over audio and video on AMD/Xilinx Kria SOMs. Provides both a no-code GUI (CLIO) and Python/Jupyter Notebook bindings, and is licensed node-locked (e.g., KV260 evaluation) for engineering teams building edge AI applications.
• MRadio Audio streams analyzer for real-time processing and analysis of audio data streams on FPGA hardware, used for applications such as broadcast monitoring and audio intelligence on edge devices.
• Mooseka Scalable audio analysis framework for processing and analyzing audio content using FPGA acceleration, supporting multimedia recognition and feature extraction at scale.
• ALOE Artificial reality framework for developing augmented reality applications on FPGA platforms, supporting AR/VR experiences accelerated on AMD/Xilinx hardware.
• Custom Software Development Services Bespoke engineering services covering software development, AR/VR technologies, machine learning, hardware acceleration on FPGA, and embedded digital signal processing. Delivered to enterprise clients in robotics, industrial automation, and healthcare under direct consulting engagements.
• Teaching and Training Courses Structured FPGA, PYNQ, and AI training courses for engineering teams and developers seeking hands-on instruction in FPGA development and edge AI deployment.
• Demo Development for Trade Shows and Product Demonstrations Custom-built hardware-accelerated demos for trade shows and product demonstrations, typically built on AMD Kria SOMs and integrated with MuseBox AI pipelines.

Segments4 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

Integration3 records

AI capability9 records

Feature6 records

Partnerships

One partnership is on record.

• AMD (formerly Xilinx)coreTechnology or IntegrationMakarenaLabs is AMD's partner specializing in FPGA-based robotic applications and AI solutions. The company develops solutions on AMD platforms including KRIA KR260, Kria K26 SOM (KV260), Zynq Ultrascale+, and Zynq-7000. MuseBox is optimized for AMD technologies supporting DPU and XRT. MakarenaLabs has developed multiple reference designs and tutorials for AMD FPGA platforms including retrogaming projects (NES, GameBoy, PacMan) and collaborative robotics demonstrations.

Scale indicators2 records

Recent moves6 records

Expansion highlights4 records

Broad incumbents

• Lattice Semiconductor: Low-power FPGA vendor with its own sensAI stack for edge AI inference on FPGAs. Competes with AMD/Xilinx in the same edge-AI market and could attract customers evaluating alternatives to the Kria ecosystem.
• AMD (Xilinx): Maker of the Kria/Zynq FPGA SoCs and DPU/XRT toolchain on which MakarenaLabs builds MuseBox. Functions as both platform partner and broad incumbent whose own reference applications and ecosystem initiatives directly compete for Kria developers' attention.
• ADLINK Technology: Edge computing hardware and platform provider with edge AI modules and software stacks for industrial and robotics applications. Comparable in serving the same industrial automation, robotics, and healthcare robotics end markets with edge AI offerings.

Emerging players

• BrainChip (Akida): Neuromorphic edge AI processor and IP company targeting low-power on-device ML inference. Competes in the same edge AI inference market with a different silicon approach, appealing to similar industrial and IoT customers.
• Nod.ai: Edge AI software optimization company building compiler/runtime tooling to deploy ML models efficiently on heterogeneous hardware including edge devices. Comparable as a software-only alternative that abstracts hardware complexity for ML developers — adjacent to MuseBox's value proposition.
• Deeplite: AI model optimization company focused on making deep learning models smaller and faster for edge deployment. Comparable as a software layer that enables edge AI on resource-constrained devices, intersecting with MuseBox's edge-AI acceleration positioning.
• SiMa.ai: Edge AI MLSoC company combining dedicated silicon with a software stack to simplify ML deployment on edge devices. Comparable as a complete platform solution targeting industrial and embedded ML workloads similar to those addressed by MuseBox.

Direct peers

• InAccel: FPGA acceleration platform and toolchain company offering runtime/SDK abstractions for deploying ML and data-processing workloads on FPGA clusters. Comparable in providing a software layer that simplifies FPGA-based acceleration for ML workloads.
• LogicTronix: FPGA/AI services and IP company building on AMD/Xilinx platforms (PYNQ, Kria, Zynq) with custom ML acceleration and edge AI deployments. Closely comparable as a small specialist offering both reference designs/IP and bespoke FPGA/ML engineering services.
• Xelera Technologies: Provides FPGA-based ML and data acceleration platforms and engineering services, similar to MakarenaLabs' combination of MuseBox productization and custom FPGA/ML services. Directly comparable in offering FPGA acceleration stacks and custom development for enterprise customers.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat4 records

Key risks5 records

Key highlights5 records

Customer concentration

• GitHubgithub.com/MakarenaLabs
• Xx.com/MakarenaLabs
• Facebookfacebook.com/makarenalabs

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about MakarenaLabs