Xecta Transforms Complex Oil Well Analysis with a Conversational AI Agent
At a glance
Xecta partnered with New Math Data to deliver a conversational AI interface for its Production Advisor, enabling users to define and iterate complex well scenarios using natural language. The agent removes form-heavy friction, accelerates scenario creation, and lowers the barrier to entry for non-expert users, positioning Xecta as a leader in user-centric energy technology.
Industry
Use Case
Workflow and process automation using generative AI, AI-augmented enterprise search, and intelligent document processing
Solution implemented
- Built a conversational AI agent that maps natural-language requests to Production Advisor scenarios
- Developed custom GraphQL tooling to translate conversational intent into precise mutations
- Implemented robust conversational state management to track and validate multi-step scenarios
- Added a linear transformation utility for on-the-fly unit conversions and parameter normalization
- Delivered a modular, extensible architecture (e.g., LangGraph) for new tools and skills
- Packaged as a Python library for containerized, production deployment (Kubernetes-ready)
The value equation
- 60% reduction in time to build common scenarios, accelerating expert workflows and deeper analysis
- 25% increase in general advisor usage, expanding accessibility and overall adoption
- 50% faster onboarding, improving time to first use and lowering barriers for new users
- A long-term foundation for proactive, AI-driven advisor capabilities and user innovation
Company Snapshot
Xecta is an integrated technology solutions provider specializing in advanced AI and machine learning combined with deep domain expertise in the energy sector. Founded in 2020 and headquartered in Houston, TX, with an office in Dubai, UAE, Xecta delivers user-friendly digital engineering solutions that fuse AI and physics to optimize production and asset management.
Location
Texas, USA and Dubai, UAE
Xecta’s AI Agent Makes Complex Well Scenarios Instantly Conversational
Xecta’s Production Advisor is a powerful analytics tool, yet building and iterating complex scenarios through traditional forms slowed expert users and raised the learning curve for newcomers. Xecta sought a faster, more intuitive way to interact with sophisticated oil and gas analytics without sacrificing rigor.
New Math Data designed and implemented a conversational AI agent that lets users describe scenarios naturally, iteratively refine parameters, and get to analysis faster, while the system ensures accuracy, completeness, and correct formatting for the underlying engine.
Problem
Despite strong capabilities, the form-based workflow introduced friction for both expert and non-expert users. The team needed to speed scenario iteration, reduce cognitive load, and lay a technical foundation for advanced AI features.
Xecta needed a scalable, production-ready solution that could:
- Accept natural-language inputs to create and modify complex scenarios
- Maintain conversational state across multi-turn interactions
- Translate domain-specific jargon into structured parameters the engine requires
- Integrate cleanly with a GraphQL backend designed for strict, pre-validated inputs
- Support unit conversions and parameter normalization reliably
- Ship as a production artifact that is easy to deploy and extend
Solution
Conversational AI for Production Advisor
A natural-language interface allows users to describe, adjust, and validate scenarios conversationally bypassing rigid forms while preserving analytical integrity.
Custom GraphQL Tooling & Schema Awareness
New Math Data built a bespoke interaction layer to read the Production Advisor’s GraphQL schema and convert conversational intent into precise mutations, ensuring validity and completeness.
Conversational State Management
The agent tracks scenario context as a structured JSON state across turns, remembering, updating, and validating parameters over multi-step dialogues.
Domain Language Understanding & Unit Conversion
A dedicated linear-transformation utility performs on-the-fly unit conversions and parameter normalization, bridging human language and the system’s required inputs.
Modular, Extensible Agent Architecture
Using a modern, modular approach (e.g., LangGraph), the agent can be extended with new tools, skills, and parameters, future-proofing the implementation as Production Advisor evolves.
Production-Ready, Containerized Delivery
Delivered as a self-contained Python library ready for containerization and deployment (including Kubernetes), enabling reliable, cloud-native operations.
Over a focused engagement, New Math Data delivered a production-ready conversational AI agent that transforms how users interact with Xecta’s Production Advisor, reducing complexity, accelerating scenario analysis, and establishing a foundation for future AI innovation.
Related Case Studies
Vertically Integrated Utility Company
Migration of data systems and applications to AWS.
Ready to Transform Your Energy or Utility Business?
See how New Math Data can transform your energy or utility business with AWS-powered innovation.