Top 50 Deep Reinforcement Learning

Phaidra develops AI-driven control systems that utilize deep reinforcement learning to optimize operations in mission-critical facilities, enhancing stability, energy efficiency, and sustainability. By replacing static, hard-coded control systems, Phaidra's technology continuously adapts and improves, significantly reducing energy consumption and CO2 emissions in industrial environments.

Predictiva develops autonomous trading platforms that utilize deep reinforcement learning algorithms to analyze market data and execute trades without human intervention. This technology addresses the challenges of human error and emotional bias in financial trading, enabling users to achieve consistent, market-beating returns across various asset classes.

Composabl provides a platform for engineers to build intelligent autonomous agents by integrating technologies such as Deep Reinforcement Learning and Machine Learning using modular building blocks. This enables organizations to automate complex industrial processes without requiring extensive coding skills, enhancing operational efficiency and reducing reliance on software developers.

This company develops general AI for robotics that learns reliable control policies directly from real-world autonomous data collection. They utilize world-model based reinforcement learning algorithms to achieve performance improvements beyond human supervision. The core value is creating robust robotic systems that minimize the need for manual teleoperation or extensive human instruction.

InstaDeep develops AI-powered decision-making systems utilizing GPU-accelerated computing, deep learning, and reinforcement learning to tackle complex challenges in industries such as logistics, energy, and biology. Their technology enhances operational efficiency and precision, enabling enterprises to make data-driven decisions in an increasingly AI-centric landscape.

Hammerhead provides a SaaS platform that uses reinforcement‑learning agents to continuously orchestrate power, cooling, and GPU scheduling in AI data centers, redirecting unused “stranded” megawatts to AI workloads in real time. By unlocking up to 30 % more compute capacity within existing power limits, the solution boosts token throughput, reduces marginal costs, and creates new revenue streams for colocation providers, AI cloud operators, and enterprise AI factories.

Develops a brain-inspired AI architecture that combines reinforcement learning and meta-learning to enable autonomous, scalable models capable of complex reasoning and planning. This approach addresses the limitations of traditional AI by balancing training costs with high performance, allowing systems to tackle demanding tasks across various domains.

May Mobility develops autonomous vehicles utilizing a Multi-Policy Decision Making (MPDM) system, a real-time reinforcement-learning AI that enables vehicles to learn and adapt to their environment every 200 milliseconds. This technology addresses the challenge of safely navigating unpredictable driving scenarios, allowing for efficient deployment of autonomous transportation solutions in diverse settings.

Aviro provides a cloud‑native platform for building and scaling training environments tailored to long‑horizon autonomous agents. Its modular composer, containerized execution engine, and persistent world state let developers create dynamic, information‑rich simulations that run thousands of concurrent instances, while integrated telemetry feeds performance data directly into reinforcement‑learning pipelines.

AgileRL provides an open-source framework for reinforcement learning that enhances training speed by up to 10 times through RLOps, supporting both single-agent and multi-agent environments. The platform utilizes evolutionary hyperparameter optimization and distributed training to enable efficient convergence on optimal performance, addressing the challenges of slow training times and complex task management in AI development.

<description>NEODE Systems supplies a modular, rugged compute stack that embeds FPGA‑based AI inference engines and containerized deep‑learning and reinforcement‑learning models directly onto missiles, UAVs, and other autonomous defense platforms. The solution uses CI/CD pipelines, open‑architecture APIs, and FIPS‑compliant communications to accelerate development cycles to weeks while meeting MIL‑STD and safety‑critical certification requirements.</description

General Intuition provides a platform that converts billions of gameplay videos into interactive 3D simulation environments for training AI agents via reinforcement learning. The resulting models gain spatial‑temporal intuition that can be transferred to robotics, autonomous navigation, and other embodied AI applications.

General Intuition provides a platform for training AI agents through rule‑based game simulations that combine perception, world‑model construction, and reinforcement learning. The system lets agents generate and test hypotheses in low‑cost simulated environments before transferring skills to real‑world robotics and autonomous systems via an open‑source API. It targets AI research labs and robotics manufacturers seeking embodied, adaptive intelligence.

Poolside is developing a foundation model specifically designed for software engineering, utilizing reinforcement learning from code execution feedback to enhance coding performance. The platform enables businesses to create custom AI models that continuously learn from their unique codebases and practices, improving developer efficiency and software quality.

Augento provides a platform for Reinforcement Learning (RL) fine-tuning to adapt foundational Large Language Models (LLMs) to specific use cases. This process relies on defining a reward function rather than collecting curated prompt-response pairs, leading to more reliable model behavior. The service automates the RL training jobs, hosts the resulting fine-tuned models, and provides API access for deployment.

Matrices provides a cloud‑native platform for building and running configurable virtual environments that simulate real‑world applications such as email, CRM, and trading systems. Developers use a drag‑and‑drop editor and a component marketplace to create sandboxed task scenarios, which agents can interact with via API for deterministic RL training at scale. The platform includes orchestration, analytics, and security features to support high‑throughput autonomous LLM agent development.

Adaptive ML develops a platform that enables companies to privately tune and deploy language models using reinforcement learning from human feedback. This technology allows businesses to enhance model performance while maintaining data privacy and optimizing outputs based on specific user metrics.

Cerebry is an AI-driven tutoring platform that provides adaptive assessments and personalized remediation for students in educational institutions. By enabling schools, publishers, and edtech companies to deliver targeted reinforcement, it enhances learning outcomes and addresses knowledge gaps effectively.

Neobotics provides an open‑source robotics platform called NeoRacer V1 that lets high schools and colleges teach and experiment with autonomous vehicle technology. The system combines affordable hardware—such as a Jetson Orin Nano, 2D LiDAR, and camera‑ready design—with a Unity‑based simulation environment for reinforcement‑learning training, enabling hands‑on learning in autonomy and machine learning.

Ineffable Intelligence builds a reinforcement‑learning based "superlearner" that acquires knowledge solely through interaction with its environment, progressing from basic motor skills to abstract reasoning without any curated human data. The platform offers a scalable, end‑to‑end RL pipeline that can be deployed on digital simulations, robotics, or other physical agents to enable continuous, self‑directed skill acquisition.

minds.ai's DeepSim platform utilizes supervised learning, reinforcement learning, and generative AI to optimize semiconductor manufacturing processes and enhance operational efficiency across all fabrication facilities. By automating software generation for hardware control and process design, it improves key performance indicators without disrupting existing workflows.

KredosAi provides an AI‑driven platform that engages consumers through their preferred communication channels to improve payment collection and reduce delinquency. The solution uses reinforcement learning with human feedback to personalize messaging, delivering faster payment pull‑forward and higher customer retention. It offers SOC 2‑compliant security, rapid deployment, and integration flexibility for enterprises in telecom, auto financing, and financial services.

Victus provides an AI-driven navigation platform that uses reinforcement‑learning models trained in high‑fidelity simulations to enable real‑time, sub‑10 ms decision making on existing drone, underwater or space robotics hardware. The hardware‑agnostic, API‑first solution runs on onboard compute or edge devices, offering scalable fleet management and customizable navigation policies without additional sensors or redesign.

Irreverent Labs builds autonomous AI assets for decentralized finance and combat sports, delivering real‑time predictive analytics and self‑executing strategies. Its flagship protocol, Boktoshi, acts as a sentient trading companion, while tools like supermodel.ai, Fight/Deck, and MechaFightClub provide crowd‑sourced DeFi trading, MMA matchup analysis, and reinforcement‑learning combat simulations through live APIs and mobile apps.

Imitation Machines provides a Robot Learning Platform that combines imitation learning and reinforcement learning to enable robots to acquire complex skills directly from human demonstrations. The platform offers an intuitive control interface, allowing users to teach robots tasks without traditional code-based programming. This service democratizes advanced robotic automation for businesses across various sectors, including manufacturing and logistics.

The startup develops a geospatial data platform that utilizes artificial intelligence, specifically reinforcement learning and computer vision, for object detection in mobile mapping data. This technology enables infrastructure and asset management professionals to efficiently monitor and analyze assets throughout their entire life cycle.

DevLand offers a collaborative coding platform with gamified AI learning experiences, allowing users to practice AI concepts like reinforcement learning and computer vision through interactive games. It also supports open-source project contributions, enabling developers to build practical skills and collaborate with a community.

Astrikos.ai offers a Smart Infra Platform that utilizes machine learning and deep reinforcement learning to analyze unstructured streaming data from operational systems in smart cities and industries. This platform provides real-time insights and predictive analytics, enabling data centers and other sectors to enhance operational efficiency and proactively manage emerging challenges.

Brick provides autonomous energy management software that leverages reinforcement learning to optimize facility HVAC and lighting systems for significant energy savings. The platform integrates seamlessly with existing hardware via protocols like BACnet and Modbus to maximize efficiency with zero operational disruption. It delivers real-time monitoring, anomaly detection, and AI-driven insights to enhance operational efficiency and support decarbonization goals.

Uno provides an AI-driven Governance, Risk, and Compliance (GRC) co-pilot that automates the analysis and remediation of vulnerabilities across multi-cloud and hybrid environments. By leveraging large language models and deep reinforcement learning, Uno enhances incident response times by up to 30 times and scales operations tenfold, enabling organizations to manage IT and security risks more effectively.

Batch Robotics offers an AI‑enabled robotics platform that combines motion‑planning, reinforcement learning, and human‑robot interaction to create autonomous robotic cells adaptable to changing tasks and environments. Its modular hardware and integrated software stack provide real‑time perception, cloud analytics, and easy integration with ERP, MES, and warehouse systems, enabling manufacturers and logistics operators to automate processes, increase throughput, and reduce labor costs without deep robotics expertise.

This platform enables users to convert proprietary software and web applications into reinforcement learning environments for agent training and evaluation. It provides a unified API endpoint compatible with OpenAI clients to access various large language models for inference testing. The infrastructure supports scalable, concurrent environment execution with low latency for rapid benchmarking and analysis.

Causal Foundry provides the kenkai platform, an adaptive AI system designed for real-time personalization and decision optimization at enterprise scale. This infrastructure leverages reinforcement learning and contextual bandits to continuously improve user engagement strategies based on high-resolution data streams. The platform delivers tailored predictions and context-aware interventions directly integrated with existing systems via governed metrics.

This company develops FuxionAI, a reinforcement learning-driven platform for real-time, multi-modal sensor fusion at the network edge. The platform integrates RF, thermal, and visual inputs to provide high-confidence target detection without cloud dependency. It supports mission-critical defense and enterprise applications like counter-UAS operations and infrastructure monitoring.

Leapwave provides an AI native platform that automates performance marketing using real‑time data ingestion and reinforcement learning.

This company develops Deep Meaning™, a novel AI framework focused on adaptability and real-world intelligence beyond conventional deep learning. Their agents build a stable world comprehension, enabling rapid generalization to new tasks with low energy consumption. Deep Meaning™ integrates into diverse applications, including robotics and AIoT, where data scarcity and speed are paramount.

WISP provides an AI-driven platform that uses deep reinforcement learning to continuously optimize traffic light sequences across city intersections, reducing vehicle idle time and emissions. The cloud‑based solution works with existing signal infrastructure, offering real‑time adaptation, travel‑time analytics, and predictive wait‑time estimates for municipal traffic agencies.

GOAT.AI develops compact language models that utilize reinforcement learning and proximal policy optimization to mitigate issues such as hallucinations and context limitations in AI-generated content. The company focuses on enhancing human-AI interactions by training smaller models that outperform larger counterparts on domain-specific tasks.

This company develops AI agents utilizing multiple LLMs and Reinforcement Learning to accelerate software and hardware verification processes. Their tools automatically generate accurate tests, stimuli, and code fixes, significantly reducing manual effort and debug cycles. The platform aims to achieve verification speedups by automating complex tasks across design verification workflows.

SI Robotics is developing a fully European humanoid robotics stack, including proprietary bionic actuators and reinforcement learning capabilities. This infrastructure is designed for dual-use applications across commercial logistics and defense sectors. The company focuses on providing a trusted, sovereign supply chain independent of non-EU technology sources.

This startup develops an artificial intelligence-powered cryptocurrency trading software that utilizes deep reinforcement learning and agent technology to generate hourly trading signals based on multiple market analysis indicators. By providing real-time dashboards for performance tracking, the software enables users to optimize trading strategies and achieve passive income with minimal execution errors.

The startup develops a cloud-based AI optimization engine that utilizes deep reinforcement learning to enhance container terminal operations. By integrating with existing Terminal Operating Systems, it reduces costs, minimizes unproductive moves, and improves visibility in yard operations.

Ludus Labs builds AI athletes trained with reinforcement learning to compete in novel sports and athletic challenges. These AI competitors, unbound by biological limitations, push the boundaries of performance and strategy in high-fidelity simulated environments.

This company applies reinforcement learning (RL) software to industrial control systems for real-time process optimization and automation. Their on-premises solution monitors process variables continuously, adapting setpoints to improve reliability and reduce operating costs. The technology allows operators to focus on maintenance and emergency response while the AI fine-tunes complex processes using historical and live data.

FractalBrain offers a continual‑learning AI platform that expands its parametric model during both training and inference, enabling permanent knowledge assimilation rather than temporary in‑context learning. Its sparse, Hebbian‑style local learning delivers orders‑of‑magnitude gains in power and data efficiency and supports unlimited context windows, while causal and model‑based reinforcement learning provide explainable, hierarchical decision‑making.

The startup develops a vision-based robotic controller that utilizes deep learning algorithms to enable robots to learn tasks by observing human actions. This technology allows businesses to automate repetitive processes, reducing labor costs and increasing operational efficiency.

This startup is developing Artificial General Intelligence (AGI) using advanced neural network architectures and reinforcement learning techniques to create systems that can perform any intellectual task a human can. The technology aims to enhance decision-making processes across various industries by providing machines with the ability to understand, learn, and adapt to complex environments.

Ryquo packages cutting‑edge NLP, autonomous agent, and reinforcement‑learning models into enterprise‑grade APIs and secure deployment options, letting organizations add advanced AI capabilities without building research infrastructure. It also runs an AI Research and Competition program that trains high‑school students in deep AI research and guides them toward academic publication.

This startup offers a cloud-based logistics management platform that uses reinforcement learning to optimize supply chain operations. Their platform integrates data, automates processes, and provides real-time visibility, enabling logistics providers and manufacturers to reduce costs and improve service.

Thoth develops AI-powered robotic solutions for advanced process automation in smart factories. Their technology enables flexible automation capable of handling uncertainty across diverse manufacturing tasks, from inspection and assembly to complex material handling. The core offering utilizes deep learning and autonomous programming to deploy adaptable robotic systems for high-mix, high-volume production environments.