The discovery of new substances follows the iterative process of Design Build Test Analyze (DMTA). Artificial intelligence has achieved intelligent design of molecular structures in the field of material discovery, much like conducting an experiment in the digital world, which we call a 'dry experiment'. However, to truly implement and continuously optimize these digital designs, it is necessary to conduct experiments and validations in the physical world, which we call "wet experiments". Real world experiments can also generate effective data to feed back into digital world models, optimizing model prediction, judgment, and design. However, in terms of experimental execution and data collection, traditional manual experiments have gradually become efficiency bottlenecks. The experimental flux is low, errors are prone to accumulate, and subjectivity is difficult to avoid. Robot chemistry, based on precision hardware and automation technology, can work continuously for 7 × 24 hours, complete thousands of experiments per day, and compress operational errors to less than 5%, greatly improving experimental efficiency and data reliability.

In the forefront of technological innovation today, the integration of artificial intelligence and automation technology is leading a profound transformation in the paradigm of material discovery research and development.Jingtai TechnologyWith its intelligent autonomous experimental platform, it seamlessly integrates data-driven AI prediction models with automated robot technology to build an efficient and autonomous material discovery system. This innovative system not only accelerates the exploration process of chemical space, but also provides a new infrastructure for research and development in industries such as pharmaceuticals and new materials, redefining the speed and efficiency boundaries of substance discovery.

The data-driven AI application paradigm

In the application of AI industry, data, as a fundamental element, constitutes the cornerstone of the entire intelligent system. Its core logic follows a positive cycle of "data → model → application → solution": high-quality data generates intelligent autonomous capabilities through training AI models, and model capabilities are transformed into specific industry application scenarios, ultimately landing into reusable solutions. This closed-loop system ensures that AI technology can continuously learn, optimize, and create value from data.

Jingtai Technology's intelligent independent experimental platform focuses ondata assetWe have established a complete data asset management system for the entire lifecycle management, from collection, governance to application. The experimental workstation, built by combining AI models and automated robots, completes intelligent experiment execution and efficient data governance at an unprecedented speed, achieving iterative closed-loop AI molecular design and reaction prediction based on algorithms and automated robot synthesis experiments, and rapidly exploring the vast chemical space.

Closed loop technology path for dry wet experiments

Jingtai Technology has built a leading intelligent autonomous experimental platform through three core technology components, forming a complete closed loop of "algorithm design → robot experiment → data feedback → model iteration". This innovative architecture breaks through the efficiency bottleneck of traditional chemical research and drives the process of substance discovery at an unprecedented speed. Through this technological architecture, the platform has achieved iterative closed-loop AI molecular design, scheme prediction, and automated robot experiments.

Application case:

Jingtai Technology Electrolyte Formula Optimization Platform

The electrolyte formula optimization platform of Jingtai Technology consists of three major modules:General artificial intelligence, proprietary large models, and autonomous laboratories.

General Artificial IntelligenceThe module mainly implements the collection of public data such as literature and patents, and the construction of a knowledge base, providing a raw material library and initial formula for project development;Proprietary large modelThe module implements functions such as molecular design, property prediction, formula optimization, and mechanism interpretation, and recommends new formulas;Independent laboratoryThe module realizes the automated and precise preparation and characterization of formulas, verifies the formulas recommended by the algorithm, and generates new experimental data to provide material support for the optimization of the algorithm model. The organic combination of three modules can efficiently achieve the expected electrolyte formula.

●Workflow of electrolyte formula optimization platform

The workflow of Jingtai Technology's electrolyte formula optimization platform is divided into six steps, namelyDetermine application scenarios, generate initial formulations, construct property prediction models, recommend new formulations using models, automate formulation preparation and characterization, and design and synthesize targeted moleculesAmong them, steps 3 to 6 are iterative to obtain the electrolyte formula that meets the requirements.

• Step OneThe project team scientists will determine the application scenarios, such as optimizing the new electrolyte formula for high voltage and wide temperature scenarios;

• Step twoBuild a specialized knowledge base and structured data processing related to the project. Firstly, a customized search template is created using prompt words, and a large language model is used to score and screen highly relevant literature and patents, constructing a specialized knowledge base; Secondly, multi-modal AI models such as LLM and OCR are used to collect project related data from literature and patents;

• Step threeBy using quantum physics simulation calculations and deep learning models, not only can relevant properties be predicted, but structures can also be generated based on expected properties for the design of new additive molecules;

• Step 4Based on the material system we have selected, we will use multi-objective Bayesian optimization and other methods to generate new formula combinations and ratios for experimental preparation and testing;

• Step 5Automated formula experiments, relying on Jingtai Technology's self-developed automated and digital electrolyte formula optimization platform, achieve precise configuration, digital recording, and characterization testing of formulas, thereby verifying the effectiveness of algorithm recommended formulas and generating new experimental data to further optimize algorithm models;

• Step SixIf it is not possible to obtain a formula that meets the target properties through different ratios in the existing compound composition space, it is necessary to expand the compound selection space to a new chemical space through directed molecular design and synthesis modules. For example, using XMolGen software from Jingtai, new molecules can be generated by replacing R groups that comply with chemical reaction rules through a commercial block library. After completing the task, you can easily perform operations such as viewing, editing, saving, exporting, and sorting. The generated molecules possess both novelty and synthesizability, and can be rapidly synthesized through automated platforms.

Driven by the acceleration of global energy transition and the upgrading of power battery performance, electrolyte, as an important part of battery development, has become a core bottleneck in breaking through key indicators such as high energy density, long cycle life, and wide temperature range adaptability through formula innovation. The traditional R&D model is difficult to meet the urgent needs of the industry for R&D efficiency and accuracy due to its reliance on experience trial and error, low data utilization, and high experimental costs. In this context, the construction of an electrolyte formula optimization platform has become an inevitable trend - it uses artificial intelligence technology to achieve structured analysis, molecular design, and property prediction of literature patent data, and relies on automated robot workstations to complete high-precision preparation and characterization of formulas, forming a closed-loop iteration of "AI prediction robot verification", effectively solving pain points such as long research and development cycles, fragmented data, and poor experimental repeatability. This "AI+robot" technology empowerment model can not only shorten the electrolyte research and development cycle, but also promote the industry's transformation from "experience driven" to "data-driven" and "intelligent driven", becoming a key infrastructure for new energy and new material enterprises to build core competitiveness.

AI x Robot Driven R&D Intelligent Transformation

Jingtai Technology's intelligent independent experimental platform integrates over 200 vertical AI models to create a chemical super brain, guiding material structure and experimental design; Efficiently completing high-throughput precision chemical execution and data governance with laboratory automation robots; The structured data collected by robots from real scenes is fed back to train AI models, making predictions more accurate and designs more optimal. Jingtai Technology's intelligent autonomous experimental platform has achieved a paradigm shift from "experience driven" to "data-driven" research through the collaboration of AI and robots, bringing dual breakthrough value to the field of material discovery.

• Technology integrator

Breaking through core technical difficulties, integrating AI algorithms, automation control, data analysis and other fields of technology

• Scene Adaptor

Provide cross functional, cross industry, and cross domain intelligent laboratory solutions to meet diverse needs

• Paradigm Innovator

Building a new paradigm of material discovery with "AI x robots" and creating a new infrastructure for industry research and development

Jingtai Technology currently has a 5000 square meter self owned intelligent independent laboratory, fully empowering internal business research and development scenarios, and accumulating technical application experience. At the same time, the industrialization achievements of the intelligent independent experimental platform have expanded to fields such as biomedicine, new energy, new materials, petrochemicals, modernization of traditional Chinese medicine, chemical automation, etc. The platform has been widely used in synthetic chemistry research scenarios covering organic synthesis, formula optimization, catalyst research and development, and electrolyte engineering.