Legal AI assistant

Siemens used Amazon Bedrock and Amazon Nova Foundation Models to streamline complex global search across 15-20 Siemens sites.Customers can enter natural-language queries and receive relevant information in seconds, instead of sifting through marketing pages to find technical documentation.An AWS Lambda function orchestrates validation, classification, summarization, and guardrail agents for the search workflow.

SiemensTech & Comms

Shorthills AI built a production-grade legal AI assistant capable of delivering complete, accurate, citation-backed answers at enterprise scale.The solution uses IBM watsonx.data to support secure hybrid retrieval over hundreds of thousands of legal documents.

Shorthills AILegal

Nexthink, a Swiss digital employee experience platform, built Nexthink Assist to let users ask natural-language questions and receive DEX insights without needing to know Nexthink Query Language (NQL).The team first used prompt engineering and RAG, then fine-tuned a specialized 7B LLaMA model on a golden dataset to improve accuracy, consistency, and coverage across the data model.They also built an automated labeling workflow using Amazon SageMaker Ground Truth and Amazon Bedrock with Claude 3.5 Sonnet as an LLM-as-a-judge, supported by S3, AWS Lambda, and Amazon SQS.

NexthinkTech & Comms

Company: LinqAlpha (institutional investors / hedge funds and asset managers). Industry: Finance (Investment research / Capital Markets). Challenge: investors need to objectively pressure-test investment theses using diverse evidence (broker reports, expert calls, SEC filings), which is slow and manually intensive while maintaining auditability and compliance. AWS Tech: Amazon Bedrock (Claude Sonnet 4.0 and Sonnet 3.7 for document parsing/VLM), Amazon EC2 (Python orchestration layer), Amazon S3 (raw document storage), Amazon RDS (structured outputs), Amazon OpenSearch Service (evidence indexing/retrieval), plus Amazon Textract integration for parsing/enrichment. Approach: LinqAlpha built the “Devil’s Advocate” generative AI research agent within a multi-agent workflow that ingests documents, decomposes thesis assertions into explicit/implicit assumptions, retrieves counter-evidence grounded to the uploaded sources, and generates structured, citation-linked critiques/JSON outputs for analyst use. Results: the agent system compresses traditional diligence cycles from days to minutes, supports evidence-linked counterarguments for auditability/traceability, and helps reduce confirmation bias by systematically uncovering blind spots before investment committee decisions.

LinqAlphaFinance

This law firm implemented a custom AI agent built on Azure AI Foundry and a custom-trained large language model (LLM) to address the challenge of time-consuming legal case analysis. The AI agent integrates with external legal databases and the firm's document management system, providing flexible orchestration and enhanced search capabilities within Microsoft 365 Copilot and the firm's internal platform. This has streamlined legal research workflows, improved productivity, and ensured secure, compliant access to case materials across platforms.

Confidential Law FirmLegal

NASA operated satellites generating over 100 petabytes of Earth Science data faced major accessibility and usability challenges for both specialists and the broader public. Traditionally, navigating geospatial datasets required specialist knowledge and intense manual effort. NASA's Office of the Chief Science Data Officer partnered with Microsoft to develop Earth Copilot, an AI conversational assistant built using Azure OpenAI Service and hosted on Azure Cloud.Earth Copilot integrates with NASA's VEDA platform, enabling plain language queries to analyze complex geospatial data—such as querying hurricane impacts or evaluating air quality trends. The agent leverages scalable AI and machine learning capabilities of Azure to handle sophisticated queries on vast geospatial datasets.This solution has opened up direct access for researchers, educators, policymakers, and the public, who can now analyze NASA Earth data with ease and efficiency. Democratizing this access supports NASA's open science mission, accelerates scientific insight, and enables faster decision making in climate, disaster response, urban planning, and agricultural applications.The project has completed its proof of concept and is undergoing testing within NASA before broad public release. Data privacy, misuse prevention, and responsible deployment of AI remain core operational considerations.

NASAPublic Sector

Ventia, a major provider of infrastructure services across Australia and New Zealand, launched a generative AI solution to improve access to internal standards, compliance, and procedural information for over 35,000 employees.The solution, RegTech GenAI, uses Microsoft Azure OpenAI Service, Azure AI Search, and Azure AI Services to power a chatbot for efficient document retrieval and semantic search.Prior to implementation, employees faced time-consuming processes for finding policy and compliance documents, elevating business risk and inefficiency.RegTech GenAI integrates directly with Ventia’s online document management system, delivering fast, accurate search results in natural language with referenced citations.A successful pilot led to a staged rollout, reaching 1,000 users initially with plans for enterprise deployment to more than 10,000 employees.Feedback from staff has been positive, citing productivity gains and improved confidence in information accuracy.Ventia’s experience is shaping further AI adoption in infrastructure operations, with ongoing digital transformation and innovation goals.The solution empowers teams to innovate quickly, creating future value and enhancing service delivery across multiple sectors.Executives highlighted that AI has become integral for optimizing operations, service excellence, and risk management.The collaboration with Microsoft included technical expertise, platform solutions, and best-practice support for responsible AI.

VentiaProfessional Services