BASF is a German multinational and one of many world’s largest chemical corporations, identified for its built-in Verbund manufacturing community, world scale, and broad portfolio spanning from primary chemical compounds to superior agricultural options. With its robust basis in analysis and growth, BASF operates throughout numerous industries whereas repeatedly driving innovation and sustainability.
One in all its key operational divisions is BASF Coatings, which makes a speciality of growing, manufacturing, and advertising and marketing superior automotive and industrial coatings, together with ornamental paints. As a pioneer in eco-efficient floor applied sciences, BASF Coatings can be on the forefront of digital transformation, leveraging AI-powered platforms to boost productiveness, innovation, reliability, and design.
In partnership with Databricks, BASF Coatings has carried out a production-ready, ruled, and business-impacting multi-agent resolution. This method not solely enhances cross-team collaboration but additionally allows smarter, sooner decision-making throughout important enterprise features — setting a benchmark for the way superior analytics and AI can drive tangible enterprise outcomes.
The Problem: Carry extra Modularity, Specialization and Management to Agent Techniques
As a company with over 11,000 workers throughout greater than 70 websites worldwide, managing the rising complexity and enhancing effectivity of cross-department digitalization is a non-trivial job. Extra particularly, turning huge, disparate organizational information into actionable insights, and enabling real-time decision-making and productiveness has turn into the important thing. Fixing this downside mattered as a result of environment friendly digital collaboration and information utilization immediately have an effect on market responsiveness, innovation pace, buyer satisfaction, and operational reliability. The stakes had been notably excessive in industries like coatings, the place agility and precision are essential amid quickly altering buyer calls for and sustainability pressures.
An agentic system – the place autonomous or semi-autonomous AI brokers proactively handle enterprise processes and information flows – was one of the best resolution as a result of it may automate coordination and evaluation duties that beforehand required intensive handbook effort. Agent techniques may empower organisations like BASF Coatings to:
- Seamlessly combine AI throughout domains, automating routine operations in gross sales, procurement, and provide chain administration.
- Present good, contextual suggestions and automate choice flows, dramatically decreasing bottlenecks and errors.
- Enhance consumer expertise by enabling “always-on” chat assistants for help, Q&A, or workflow integration throughout departments.
- Drive adoption of on a regular basis AI instruments company-wide, making advanced digital capabilities accessible to enterprise stakeholders and fostering information literacy.
As highly effective as an agent may very well be, as we develop these techniques, they may develop extra advanced over time, making them more durable to handle and scale. For instance, an agent can have too many instruments at its disposal and make poor selections about which instrument to name subsequent, additionally the context grows too advanced for a single agent to maintain monitor of. There’s a want for a number of specialization areas within the system (e.g. supervisor, area orchestration, material professional, and so forth.)
One other technique to view the problem is thru the variety of information that kinds the agent system’s data base. Many people are already conversant in RAG (Retrieval-Augmented Technology), a method that mixes giant language fashions (LLMs) with real-time information retrieval to enhance response accuracy and relevance. Nonetheless, RAG techniques are primarily designed to deal with unstructured information – Akin to paperwork, internet pages, PDFs, or different types of free textual content – relatively than structured tables with predefined fields and relationships. When working with structured information, Textual content-to-SQL is the most typical method for pure language analytics. Nonetheless, it typically depends on pre-defined instance SQL queries and lacks built-in mechanisms for information governance and permission management.
The Resolution: An Finish-to-Finish Multi-Agent Supervisor for Structured and Unstructured Info
To deal with these challenges, we suggest breaking our software into a number of smaller, unbiased brokers and composing them right into a multi-agent system. This method will comply with a supervisor sample that coordinates the specialist brokers – particularly, Genie brokers and function-calling brokers – which work together with the Databricks Vector Retailer Retrieval instrument.
AI/BI Genie, one of the vital common options inside Databricks, is designed to make structured information similar to Delta tables and views immediately accessible to enterprise customers by leveraging pure language interfaces. It makes use of metadata from Unity Catalog, similar to desk descriptions, PK/FK relationships, and column names/descriptions. This metadata guides Genie in parsing consumer questions, setting up correct SQL, and delivering contextually related solutions – serving to to mitigate errors or hallucinations. As well as, Genie authors can improve the area by domestically enhancing metadata, defining joins, including synonyms, and curating BASF-specific directions. This enables information stewards to actively handle and keep the standard of their Genie areas thus contributing on to the agent system with their invaluable enterprise area data.
To ease the usage of Genie inside agent orchestration frameworks, there are frameworks supporting devoted Python wrappers for constructing Genie brokers (test right here for reference). As well as, Databricks product group options instance notebooks that stroll our customers by way of establishing a multi-agent system utilizing Mosaic AI Agent Framework along with Genie. These examples leverage LangGraph (an open-source agent orchestration library) and show the best way to compose workflows the place Genie is one specialised agent amongst a number of.
An outline of our structure is as follows. We undertake Databricks’ Mosaic AI framework to simplify the complexities of managing AI agent lifecycles, providing instruments and speedy multi-agent coordination prototyping, rigorous analysis, and efficient real-time operational monitoring. Notably, we additionally combine the deployed supervisor endpoint with Microsoft Groups for real-time agent execution, and make AI-powered insights available to all sorts of customers, together with enterprise stakeholders who’re much less conversant in information platforms – by embedding conversational deployment endpoints immediately throughout the Groups interface. Clear, reusable accelerators exist for provisioning cloud assets (Azure Bot Service, App Service) and connecting endpoints to Groups.
Actual Enterprise Impression
Whereas BASF Coatings is growing AI brokers that may improve its enterprise processes, the primary touchdown zone challenge, Marketmind, focuses on the Gross sales & Advertising division. The use case allows superior quantitative and qualitative evaluation by consolidating inner Salesforce buyer go to stories and market consumption insights with exterior market developments together with S&P 500 information. A few of this information is already processed and out there within the type of Delta tables and views, whereas the remaining exists as free-text information and PDF paperwork, every arriving at completely different speeds and being up to date at various frequencies. Moreover, the information is managed by completely different groups and stewards. For instance, structured tables are primarily supplied by BASF’s central Enterprise Information Lake (EDL) group, with Gross sales & Advertising enterprise specialists enriching them with domain-specific metadata. In distinction, unstructured information is primarily processed by way of code-first ETL pipelines developed and maintained by the Coatings Information & AI workplace group.
Given the complexity of the information panorama, we adopted the multi-agent supervisor structure for the Marketmind challenge and used the template pocket book as our place to begin. We created a Genie area for structured information, enriching it with curated tables, detailed column descriptions, Genie-local be a part of relationships, and worth sampling. To enhance accuracy, we added SQL examples and clear directions to information Genie’s responses, and we carried out common Benchmark checks as new information got here in to judge its total efficiency.
For unstructured information similar to Salesforce go to stories and market information, we constructed vector search indices for every supply utilizing embeddings to allow context-aware similarity search. We then created Unity Catalog features that wrap Mosaic AI Vector Search queries, making certain enterprise-ready governance, discoverability, and computerized MLflow tracing. Lastly, we developed a perform tool-calling agent that invokes vector retrieval instruments to deal with task-specific requests handed alongside by the supervisor.
Our Marketmind challenge started its scoping part in April this 12 months, adopted by a 5–6 week proof of idea (PoC). We then moved into the total implementation part, accompanied by technical upskilling workshops, structure opinions, and product and have discussions with the Databricks’ Mosaic AI product group. We carried out a one‑month pilot with 25 key customers, and are actually within the ultimate refinement stage earlier than go‑stay and rolling out to North America by the tip of October . As soon as launched, greater than 1,000 gross sales representatives worldwide might be utilizing Marketmind, with inputs up to date continuously.
Marketmind is already altering how BASF Coatings’ gross sales groups put together, have interaction, and comply with up with their prospects. As a substitute of trying to find leads by way of scattered notes and folders, gross sales representatives obtain personalised notifications alongside advised actions and techniques primarily based on present occasions out there. If additional info is required, Marketmind presents the choice to dig deeper into the underlying information and stories utilizing an easy-to-use chat interface. The screenshot beneath illustrates this shift. Alerts from the market are introduced in an actionable, conversational interface inside Microsoft Groups, so Coating’s gross sales group can shift their focus from “What occurred?” to “What ought to I do subsequent?” with out switching instruments.
As proven above, gross sales groups can’t solely ask ad-hoc inquiries to the Marketmind chatbot immediately in Groups, but additionally obtain proactive adaptive playing cards with the most recent market developments on a weekly foundation. Customers can discover matters of curiosity in higher element by clicking the connected URL, which redirects them to the unique information supply. To additional improve the agent’s high quality, we have now additionally built-in a voting mechanism that enables customers to rapidly give a thumbs up or down, or present extra detailed written suggestions within the backside area. This suggestions is captured within the mannequin inference desk and built-in with the present payload information.
“Marketmind turns our area interactions into well timed, AI-driven actions—nudging good follow-ups, surfacing related alternatives, and connecting friends going through comparable challenges. The outcome: sooner prep, sharper buyer conversations, and extra time promoting the place it counts.” — Adrian Fierro, Head of International Market Intelligence at BASF Coatings
Why It Labored
Multi-agent structure with Genie as an agent presents a number of vital benefits for enterprises like BASF that look to leverage AI successfully of their enterprise contexts. We conclude the important thing power into the next points:
Specialised agent capabilities with excessive scalability and modularity: inside a multi-agent system, varied brokers can concentrate on their particular domains or duties, enabling deeper experience in dealing with numerous queries and datasets. Furthermore, organisations like BASF can increase their gateway to AI options with an structure that enables every enterprise division to function independently whereas being centrally orchestrated. This modular design helps handle complexity over time.
Enhanced collaboration and improved consumer expertise: brokers can share info and context with each other, permitting for extra complete responses that combine information from a number of sources. This facilitates smarter, sooner decision-making throughout varied enterprise features. By integrating AI endpoints to MSFT Groups as a chat interface, we enable customers to work together with brokers utilizing pure language, making it extra accessible to non-technical stakeholders.
Governance and compliance: Defending private and buyer information is the Commented basis of Marketmind and stays our highest precedence. Each interplay is constructed on strict compliance with BASF’s information safety requirements, leveraging Databricks’ enterprise-grade governance capabilities similar to Unity Catalog for fine-grained entry management, lineage monitoring, and auditability. This ensures that whereas Marketmind accelerates insights and actions, it does so inside a safe, clear, and absolutely ruled atmosphere.
Shut group work between BASF, Databricks and companions: From challenge begin, BASF Coatings, Databricks account and product groups, and companion Accenture proactively engaged in workshops,. which helped align enterprise targets, technical necessities, and product imaginative and prescient, setting a robust basis for profitable implementation. Proper on time, hands-on periods created speedy suggestions loops. Professional steerage was repeatedly supplied by Databricks product group, serving to to customise the answer for the advanced, evolving wants of BASF and making certain enterprise-grade high quality.
Wanting Ahead: Multi-Layered Orchestration and Agent Bricks
With the success of the Marketmind multi-agent supervisor resolution, the corporate is now increasing the enterprise affect throughout broader operations, together with Provide Chain, Procurement, Chemetall (Floor Expertise subsidiary), and Folks & Tradition. Along with our product group, we’re exploring a extra scalable multi-layered structure, the place every division operates its personal multi-agent supervisor, whereas a higher-level Coatings-wide orchestrator serves all customers. This hierarchical system – a “supervisor of supervisors” – strikes the suitable steadiness: it allows division-scoped information and power entry management, preserves flexibility in agent growth, and helps a Coatings-wide “Ask Me Something” functionality.
One in all our future enhancement targets is the adoption of Agent Bricks, launched this 12 months on the Information & AI Summit. Whereas our present Mosaic AI–primarily based resolution helps multi-agent orchestration, it stays code-first and requires a extra hands-on method with added complexity in deployment and administration. Agent Bricks presents a streamlined technique to construct and optimize domain-specific, high-quality AI agent techniques for widespread use instances, together with multi-agent setups. With options similar to computerized optimization, price and high quality effectivity, and user-driven suggestions mechanisms, it simplifies agent implementation and permits groups to concentrate on core challenges – information, metrics, and problem-solving. Though we have now not but been in a position to absolutely take a look at its capabilities because of restricted regional availability, we view Agent Bricks as a visionary path and plan to allow integration as soon as it turns into accessible, accelerating division-specific multi-agent supervisor growth.