Hapag-Lloyd stands as one of many world’s main liner transport firms, working a contemporary fleet of 313 container ships with a complete transport capability of two.5 million TEU (Twenty-foot Equal Unit—a regular unit of measurement for cargo capability in container transport). The corporate maintains a container capability of three.7 million TEU, which incorporates one of many trade’s largest and most trendy fleets of reefer containers. With roughly 14,000 workers within the Liner Transport Phase and greater than 400 workplaces unfold throughout 140 nations, Hapag-Lloyd maintains a strong world presence. By 133 liner providers worldwide, we facilitate dependable connections between greater than 600 ports throughout the continents.
The corporate’s Digital Buyer Expertise and Engineering workforce, distributed between Hamburg and Gdańsk, drives digital innovation by growing and sustaining customer-facing internet and cell merchandise.
Over the previous years, the Digital Buyer Expertise and Engineering workforce has developed from a delivery-focused channel into a real digital product driver, with robust buyer focus, engineering excellence, and measurable enterprise influence. We take end-to-end possession of our digital merchandise, combining customer-centric innovation with engineering craft to instantly help development and enterprise outcomes. Constructing on a contemporary, independently owned tech stack and a excessive stage of engineering maturity, we’re dedicated to staying on the forefront of expertise. Now, we’re taking the following step by shifting towards turning into AI-native, investing closely in synthetic intelligence as a core functionality. This journey is about amplifying highly effective engineering with AI to construct smarter merchandise, sooner innovation, and higher buyer worth.
Understanding consumer influence.
Up to now, our buyer suggestions evaluation course of had largely been guide and reactive. Particularly forward of evaluation ceremonies, manually analyzing buyer suggestions might take hours, typically days, when tons of of scores and feedback wanted to be reviewed. Each two weeks, Product Managers exported buyer suggestions knowledge as CSV information, learn by means of giant volumes of feedback, and manually categorized sentiment and themes. Though this work was precious and deeply linked to product selections, it was additionally repetitive, time-consuming, and troublesome to scale, limiting flexibility at any time when sooner or deeper insights have been wanted.
With our generative AI answer, we basically modified this strategy. As a substitute of manually aggregating and deciphering suggestions, we now automate your entire workflow: gathering buyer feedback, extracting sentiment, figuring out themes, and surfacing actionable insights. Product Managers and groups can focus much less on operational evaluation and extra on technique, innovation, and creating distinctive consumer experiences.
On this put up, we stroll you thru our generative AI–powered suggestions evaluation answer constructed utilizing Amazon Bedrock, Elasticsearch, and open-source frameworks like LangChain and LangGraph. Amazon Bedrock is a completely managed service that provides a alternative of high-performing basis fashions from main AI firms resembling AI21 Labs, Anthropic, Cohere, DeepSeek, Luma, Meta, Mistral AI and Amazon by means of a single API, together with a broad set of capabilities it’s essential construct generative AI purposes with safety, privateness, and accountable AI. With this answer, you may mechanically ingest buyer feedback, generate wealthy summaries, and ship focused insights. This enables our product groups to make sooner, smarter selections and drive steady enchancment.
We stroll you thru the structure and implementation of this answer, demonstrating how utilizing generative AI foundations, resembling orchestration, knowledge administration, safety, and privateness, allowed us to quickly construct a scalable, production-ready suggestions processing pipeline.
Resolution overview
AWS structure for automated suggestions processing and evaluation, using Lambda features for knowledge ingestion from Amazon S3, Amazon Bedrock for AI-powered insights accessed by stakeholders through Amazon ECS, and Elasticsearch for indexing and querying suggestions knowledge with e-mail notifications through SES.
The answer is constructed on AWS structure designed to handle these challenges by means of scalability, maintainability, and safety. It’s deployed through the use of AWS CloudFormation.
- Steady & Quarterly Suggestions Assortment
- Our internet and cell purposes serve tons of of hundreds of consumers every month.
- Customers can depart a score plus textual content feedback, serving to us perceive consumer expertise and enhance providers.
- Each day Suggestions Ingestion & Processing
- A AWS Lambda perform runs as soon as per day to fetch the brand new suggestions entries.
- We use Amazon Bedrock to classify sentiment (optimistic, adverse, blended, or impartial) for every open remark, streamlining downstream evaluation.
- Processed information are listed in Amazon OpenSearch Service, serving each as our full-text search engine and vector database.
- Interactive suggestions exploration through OpenSearch Service
- Stakeholders can entry real-time suggestions insights by means of OpenSearch Dashboards, giving them a hen’s-eye view of consumer sentiment, scores, and developments over time.
- Beginning with high-level visualizations, resembling sentiment distribution, score scores, and suggestions quantity, customers can drill down into particular purposes, options, and even particular person feedback.
- Dashboards help filtering by time interval, remark sentiment, product model, and extra, enabling focused root trigger evaluation.
- For instance, a Product Supervisor can visualize how sentiment round a current app replace modified week over week, or zoom into adverse feedback mentioning a particular characteristic.
- AI-Powered Inside Chatbot
- Our stakeholder-facing chatbot queries the OpenSearch index as its data base.
- We use Bedrock Guardrails, to implement security and reliability and ensure responses align with our model and compliance requirements.
- Product managers and help groups can ask natural-language questions, for instance, “What ache factors do prospects point out most frequently?” and obtain immediate, context-rich solutions.
- Biweekly Insights Report
- Each two weeks, a second Lambda perform aggregates and analyzes the most recent suggestions developments.
- It generates a concise report with key metrics, highlights, and sentiment breakdowns.
- The report is mechanically delivered to our Product Managers and Product Homeowners, feeding instantly into dash planning and roadmap discussions.
Generative AI Orchestration
Orchestration is a core basis of our answer, as a result of generative AI workflows usually contain a number of steps that have to be coordinated. In our pipeline, knowledge ingestion and processing steps, resembling sentiment evaluation, embedding technology, and indexing, are orchestrated utilizing LangChain, which gives modular, reusable elements for calling fashions, reworking knowledge, and integrating with exterior programs like Amazon OpenSearch Service. For our inner chatbot, we depend on LangGraph to implement a multi-agent structure. Every assistant is outlined declaratively in LangGraph, encapsulating its personal logic and instruments. This design makes assistants versatile and composable: as an alternative of inflexible step-by-step flows, we use an agent-based strategy the place an LLM selects the appropriate instruments and actions dynamically to reply consumer queries. This offers product managers and help groups a pure, interactive solution to discover suggestions and associated operational insights.
Integration with Amazon Bedrock fashions is easy utilizing LangChain’s native help. For instance, our AI-powered inner chatbot makes use of the Claude Sonnet 4.6 mannequin through Amazon Bedrock. We selected Claude Sonnet 4.6 as a result of it affords frontier efficiency throughout coding and agentic workflows. The mannequin excels in multi-turn conversational exchanges and agentic workflows, making it ideally suited for our inner chatbot that requires dependable efficiency throughout single and multi-turn interactions with stakeholders. With its exact workflow administration capabilities and skill to serve in each lead agent and subagent roles, Claude Sonnet 4.6 delivers the constant conversational high quality our product managers and help groups want when exploring suggestions insights at scale. Moreover, we leverage geographic Cross-Area Inference Service (CRIS) endpoint to seamlessly handle unplanned visitors bursts by distributing compute throughout a number of EU AWS Areas. This cross-region functionality ensures our suggestions processing pipeline stays resilient throughout peak utilization intervals whereas sustaining constant efficiency for our world stakeholder base. The mannequin is configured with guardrails utilized instantly by means of LangChain configuration:
from langchain_aws import ChatBedrockConverse
def get_chatbot_model():
return ChatBedrockConverse(
consumer=session.consumer(“bedrock-runtime”, region_name=”eu-central-1″, config=config),
mannequin=”eu.anthropic.claude-sonnet-4-6″, guardrail_config={
“guardrailIdentifier”: settings.GUARDRAIL_ID,
“guardrailVersion”: “DRAFT”,
“hint”: “enabled”
}
)
Information Administration
An AWS Lambda perform runs as soon as per day to fetch the brand new suggestions entries from the suggestions repository into Amazon S3, after which the info is categorized with semantic detection by means of Amazon Bedrock. The information is then listed in Amazon OpenSearch Service, serving each as our full-text search engine and vector database.
Accountable AI
To responsibly use the answer, we implement safeguards utilizing Amazon Bedrock Guardrails. This enables us to connect Amazon Bedrock Guardrails to an AI interplay and implement content material moderation insurance policies and ensure responses align with our model and compliance requirements.
Utilizing AWS CloudFormation, we outline guardrail insurance policies as infrastructure-as-code, offering examples of configurations to assist block dangerous content material.
Guardrails as Code: CloudFormation
ChatbotGuardrail:
Sort: AWS::Bedrock::Guardrail
Properties:
Identify: guardrail
Description: Fundamental guardrail to dam violence and dangerous content material.
BlockedInputMessaging: “Enter blocked by security coverage.”
BlockedOutputsMessaging: “Response blocked by security coverage.”
WordPolicyConfig:
ManagedWordListsConfig:
– Sort: PROFANITY
ContentPolicyConfig:
FiltersConfig:
– Sort: HATE
InputStrength: HIGH
OutputStrength: HIGH
OutputAction: BLOCK
– Sort: INSULTS
InputStrength: HIGH
OutputStrength: HIGH
OutputAction: BLOCK
– Sort: SEXUAL
InputStrength: HIGH
OutputStrength: HIGH
OutputAction: BLOCK
– Sort: VIOLENCE
InputStrength: HIGH
OutputStrength: HIGH
OutputAction: BLOCK
– Sort: MISCONDUCT
InputStrength: HIGH
OutputStrength: HIGH
OutputAction: BLOCK
– Sort: PROMPT_ATTACK
InputStrength: NONE
OutputStrength: NONE
OutputAction: BLOCK
Programmatic Enter Validation
We additionally use Amazon Bedrock Guardrails independently to validate uncooked consumer enter earlier than passing it to the LLM, serving to forestall immediate injection and different misuse:
def validate_question_with_guardrail(query: str, user_data: UserData) -> bool:
consumer = boto3.consumer(‘bedrock-runtime’)
response = consumer.apply_guardrail(
guardrailIdentifier=settings.GUARDRAIL_ID,
guardrailVersion=’DRAFT’,
supply=”INPUT”,
content material=[{‘text’: {‘text’: question}}]
)
if response.get(“motion”) == “GUARDRAIL_INTERVENED”:
print(json.dumps(response, indent=4))
print(
f”Immediate was blocked. user_id=[{user_data.user_id}] query=[{question}]”
)
return False
return True
With this setup, we now have created a safer, scalable, and explainable pipeline that places Generative AI to work, responsibly and successfully, throughout our product suggestions lifecycle.
Monitoring
We monitor the components of the applying utilizing Amazon CloudWatch, which collects uncooked knowledge and processes it into readable, close to real-time metrics. We enabled mannequin invocation logging to gather invocation logs, mannequin enter knowledge, and mannequin output knowledge for the invocations, enabling assortment of full request knowledge, response knowledge, and metadata related to the calls. Amazon Bedrock additionally integrates with AWS CloudTrail, which captures API requires Amazon Bedrock as occasions. This generates insights that you should utilize to optimize the purposes additional like enhancing response latency or lowering prices.
Subsequent Steps
The answer processes over 15,000 suggestions objects per 30 days with an accuracy of 95% for sentiment classification on a labeled check dataset. As a substitute of spending hours reviewing uncooked suggestions, groups can now obtain clear, structured summaries in seconds that spotlight an important subjects and recurring indicators. This helps them transfer from data to motion a lot sooner, making selections inside days reasonably than weeks. Over time, the experiences helped us perceive not solely when sentiment improved, but in addition why it didn’t. By repeatedly monitoring buyer suggestions, we will react shortly to early indicators, modify selections, and proper course when wanted. In a number of areas, actions taken primarily based on these insights have already resulted in additional optimistic feedback and a noticeable discount in adverse suggestions. A robust instance is the “Preview” performance in Transport Directions. This characteristic was prioritized instantly in response to a excessive quantity of adverse consumer suggestions highlighting the dearth of a preview functionality. After its launch, AI-driven experiences allowed us to trace consumer reactions intimately. Suggestions associated particularly to this characteristic confirmed that the beforehand frequent requests for preview functionality have been successfully resolved, demonstrating that the core consumer want had been efficiently addressed. On the identical time, broader suggestions continued to floor different areas for enchancment throughout the applying. AI insights additionally guided future characteristic planning and prioritization. Based mostly on consumer feedback, we created new OpenSearch Service-based dashboards that assist groups shortly confirm and analyze points reported by customers. One other instance is the flexibility to add cargo knowledge through Excel information, a repeatedly requested characteristic highlighted by AI suggestions. This performance is now absolutely out there and is predicted to considerably scale back guide effort, notably for big shipments. Throughout evaluation classes, the stakeholders can now see high optimistic and adverse feedback in actual time, alongside AI-generated suggestions, creating a much more knowledgeable and productive dialogue.
This suggestions evaluation answer is one instance of how we’re making use of generative AI throughout our processes, and it marks the start, not the tip, of our AI-native journey. Below our AI-Native Umbrella Program, which serves as a single supply of reality for AI adoption, our subsequent focus is to determine a shared, sturdy AI basis with Amazon Bedrock. By offering standardized infrastructure, safety, and guardrails, we purpose to allow each function within the division, engineering, product and supply (PM, PO, SM), UX/design, and operations/help, to create their very own AI “areas” safely and independently whereas gaining access to the very best in-class basis fashions. This setup is designed to decrease the barrier to experimentation, streamline discovery, and encourage hands-on exploration of generative AI use instances in day-to-day work. In doing so, we assist groups transfer sooner from concepts to influence, whereas sustaining consistency, duty, and scalability throughout the AI initiatives.
If you wish to scale your generative AI purposes, you may get began by studying this Architect a mature generative AI basis on AWS that dives deeper on the assorted foundational elements that assist speed up the end-to-end generative AI software lifecycle.
Concerning the authors
Aamna Najmi
Aamna is a Senior Specialist Options Architect for Generative AI specializing in Anthropic fashions and operationalizing and governing generative AI programs at scale on Amazon Bedrock. She helps ISVs clear up their challenges, embrace innovation, and create new enterprise alternatives with Amazon Bedrock. In her spare time, she pursues her ardour of experimenting with meals and discovering new locations.
Anna Rysicka
Anna is a Software program Engineer Staff Chief at Hapag-Lloyd with over 10 years of expertise within the expertise trade. She works with the Documentation & Customs Staff, specializing in cargo documentation programs together with Transport Directions and Invoice of Lading options. She focuses on trendy frontend architectures, TypeScript/Vue.js, and consumer expertise design, with a ardour for problem-solving and leveraging AI as a collaborative software for activity administration and productiveness. As a part of the DXE AI Native initiative, Anna guides groups into the AI period with instruments and motivation to boost productiveness. In her spare time, she experiments with AI tasks, enjoys portray, and loves touring.
Grzegorz Kaczor
Grzegorz is a Cloud Architect at Hapag-Lloyd and a expertise fanatic with over 18 years of expertise within the tech trade. He focuses on leveraging superior applied sciences to ship revolutionary options for organizations. His background spans serverless architectures, safety, observability, and compliance posture administration. He’s at present exploring generative AI and its sensible purposes in enterprise environments. In his spare time, he enjoys studying, staying energetic, and spending time along with his daughters.
