Back

Gen AI powered benchmarking and data analytics automation platform for Proxure

Deveshi Dabbawala

April 28, 2026
Table of contents

Proxure is a legal tech company that provides an AI driven data analytics automation and benchmarking platform for legal spend management. It helps organizations analyze costs, compare performance with market benchmarks, and generate insights using natural language queries. Built on the RIVER system, the platform combines structured analytics with conversational AI to simplify data access and improve decision making.

Problem: fragmented analytics and limited accessibility

Legal and procurement teams relied heavily on static dashboards and manual querying methods to analyze legal spend data. This limited the adoption of data analytics automation, as most workflows required predefined filters or SQL expertise. Users found it difficult to access insights quickly, which increased dependency on analysts and slowed decision making. The lack of flexible querying prevented users from exploring data dynamically or automating repetitive analysis tasks.

Benchmarking against market data was also manual and time consuming, reducing accuracy and limiting the effectiveness of analytics workflows. The system lacked contextual continuity, which meant users could not refine queries or build on previous insights. This prevented efficient data analytics automation and led to fragmented workflows. These gaps reduced productivity, increased operational effort, and limited the overall value derived from analytics systems.

Solution: data analytics automation with conversational intelligence

GoML built a unified platform for Proxure using its AI Data Analytics Accelerator, combining benchmarking APIs with conversational AI and data analytics automation. The system enables users to query data using natural language and automatically generate insights, visualizations, and comparisons, while accelerating development and ensuring scalable, consistent analytics workflows.

Core system architecture

  • Benchmark API for structured analytics  
  • Chat Agent API for conversational queries  
  • Real time data access and CSV export  
  • LLM driven data analytics automation  

Conversational AI and context

  • Natural language queries converted into SQL  
  • Outputs include insights, visualizations, and session continuity  
  • Supports client, market, and comparative queries  
  • Enables follow up queries without restarting  

Benchmarking and analytics engine

  • Covers key areas: overview, rates, leverage, timekeepers, fees  
  • Runs parallel computations for faster results  
  • Outputs include comparisons, gaps, and insights  
  • Enables scalable data analytics automation  

Data processing

  • Real-time data from PostgreSQL RDS with automated responses  
  • LLM-based query understanding and SQL generation  
  • Data extraction from Databricks SQL tables  
  • Sensitive data anonymization using SHA-256  
  • Data enrichment and categorization using AWS Bedrock and Azure API

User experience

  • Natural language triggers analytics workflows  
  • Dynamic visualizations and instant CSV export  
  • Session continuity for iterative analysis  

Infrastructure

  • PostgreSQL RDS and LLM integration  
  • API driven architecture  
  • Session tracking and system monitoring  

Quality assurance

  • Input validation and error handling  
  • Secure authentication  
  • SQL validation  
  • End to end testing

Impact

  • 40-60% faster insights  
  • 50-70% less manual dashboards and SQL  
  • 30-45% better decision making  
  • 2x-3x higher automation adoption  
  • Supports 3x-5x growth

About

Location 

Calgary, Alberta 

Tech stack 

AWS, Amazon Bedrock, Lambda, FastAPI, PostgreSQL RDS, S3, CloudWatch 

Before Gen AI and after Gen AI

Area 

Before Gen AI 

After Gen AI 

Analytics 

Manual and dashboard driven 

Data analytics automation enabled 

Query handling 

Static and limited 

Dynamic natural language queries 

Context 

Not supported 

Session aware workflows 

Benchmarking 

Manual comparison 

Automated benchmarking 

User experience 

Fragmented 

Unified and automated 

Results 

Static reports 

Real time automated insights 

“Proxure transformed its analytics and benchmarking workflows into a scalable data analytics automation system that improves speed, accuracy, and decision making.”

Prashanna Rao, Head of Engineering, GoML

Key takeaways

Common challenges

  • Manual analytics workflows slow down teams
  • Limited adoption of automation
  • Difficulty in scaling benchmarking

Practical guidance

  • Adopt data analytics automation for faster insights
  • Enable natural language querying
  • Combine benchmarking with automation
  • Use session context to improve workflows
  • Start with high impact analytics use cases

Ready to build data analytics automation platforms

Partner with GoML to build scalable Gen AI platforms for benchmarking, conversational AI, and data analytics automation, delivering faster insights with AI Matic.

Outcomes

40-60%
Faster insights
2-3x
Higher automation adoption
30-45%
Better decision making