Build the Future with AI You Understand and Control
We help companies design, validate, and deploy practical AI solutions that solve real problems — fast. From business case to working APIs, we make AI adoption tangible.
Our Services
AI & Data Strategy
Build your roadmap to AI maturity with actionable, investment-ready plans.
Learn More ↓Forecasting & Anomaly Detection APIs
Use DriftMind’s adaptive engine to make sense of time series data at scale.
Learn More ↓Custom Prototypes & MVPs
We rapidly turn use cases into proof-of-concept solutions you can test with your team.
Learn More ↓Conversational Interfaces
Chat with your data – literally. Our AI agents generate SQL, visuals, and explain them to you.
Learn More ↓AI & Data Strategy
Despite years of experimentation, most telecom and data-heavy enterprises still struggle to scale AI meaningfully. The issue isn’t hype fatigue, it’s that the path to value remains unclear. Too many initiatives focus on technology instead of outcomes, or leap into development without proper alignment, readiness, or feasibility. At Thingbook, we help our customers, at the CxO & VP levels, to adopt the organizational patterns that distinguish successful AI adopters from those stuck in endless experimentation. This service draws on over two decades of consulting and delivery experience.
What’s included:
- Executive-level advisory to define a clear, organization-wide AI strategy.
- Design of investment-ready business cases grounded in real feasibility and ROI.
- AI maturity assessments across six key pillars: strategy, data, tech, people, vendors, governance.
- Creation of an AI ecosystem with the right blend of internal talent and external partners.
- Vendor selection frameworks to avoid lock-in and AI-washing traps.
Our Differentiator:
We’ve helped telecoms and technology firms move beyond the hype to build real, production-ready AI capabilities. Unlike many advisors, we combine deep technical acumen with strategic insight. We don’t just talk AI, we help you architect the team, the partnerships, and the path to value. Our frameworks are used by top-tier CTOs to:
- Move from intuition-based to data-driven decision-making.
- De-risk AI investment by aligning with maturity and talent realities.
- Identify use cases that deliver tangible outcomes — not just PoCs.
Our Customers:
Client names have been anonymized or withheld to respect confidentiality agreements
Forecasting & Anomaly Detection APIs
We help organizations design, prototype, and operationalize advanced **Anomaly Detection, Real-Time Pattern Matching and Forecasting systems**. These solutions are available as scalable, enterprise-grade APIs, allowing for transparent integration of insights directly into your existing data pipelines, dashboards, and custom applications. Our in-house **DriftMind** engine provides adaptive, accurate analysis of time series data at scale.
What’s included (Consulting & Implementation):
- Use case discovery, feasibility assessment, and risk mitigation planning.
- Architecture and implementation of end-to-end Forecasting and Anomaly Detection pipelines.
- Toolchain selection based on client needs, including our in-house DriftMind engine or open frameworks like Prophet, PyOD, and Scikit-learn.
- Packaging into APIs, dashboards, or Excel/BI integrations.
Use Cases We've Delivered:
- Insurance: Fraud detection in policy claims using multi-perspective anomaly scoring (VHI, Ireland).
- Finance: Behavioral anomaly detection for anti-money laundering workflows (Banorte, Mexico).
- Telecom: Predictive maintenance from network elements or cells (MYCOM UK, T-Mobile US).
- IoT / Smart Energy: Predictive maintenance for wind farms (Hydro Quebec, Canada).
- Telecom: Network intrusion detection (Globe, Philippines).
- SaaS & Telecom: Behavioral churn prediction from drift and usage collapse signals (AT&T US, COX US, VIVO Brazil).
Enterprise-Grade APIs & Features:
Integrate our core ML capabilities directly into your workflows.
- **`/forecast`**: High-speed, Self Adapting forecasting endpoint.
- **`/anomaly`**: Real-Time Anomaly Score based on multi-perspective inputs.
- **`/match`**: Pattern matching for sequences or templates.
- SaaS metering & quotas per customer.
- Full Swagger/OpenAPI documentation.
- Pay-as-you-grow usage plans.
Tech Stack (flexible):
DriftMind, Prophet, PyOD, Scikit-learn, XGBoost, Redis, PostgreSQL, Kubernetes, Kafka.
Our Customers:
We’ve supported Complex Anomaly Detection projects for leading organizations across telecom, finance, energy, and software sectors — including MYCOM OSI, Banorte, T-Mobile, Telefónica, VHI, Sprint, AT&T, COX, VIVO, Hydro Quebec, and others.
Custom Prototypes & MVPs
At Thingbook, we believe that the most effective way to bridge the gap between vision and execution is through focused, lightweight prototypes. These MVPs are not just technical tests, they’re strategic experiments designed to validate both commercial traction and technical feasibility. By building early, measurable artifacts, we help clients avoid premature scaling, de-risk investments, and gain internal sponsorship, often revealing that organizational misalignment, not data quality, is the real blocker to AI success.
What’s included:
- Personalized recommendation engines.
- Anomaly detectors trained on your own data.
- Conversational AI interfaces linked to your DBs.
- Output: working PoCs, API endpoints, dashboards.
Our Methodology:
- **1. Define:** We work with you to define the core use case and success KPIs.
- **2. Iterate:** We rapidly build and refine a minimum viable solution.
- **3. Deploy:** We help you deploy on your infrastructure or via Thingbook's platform.
Our Customers:
Since our foundation, we’ve partnered with a diverse range of enterprise customers, from Tier 1 Telecom Operators and SaaS platforms to insurers and energy providers, helping them accelerate their AI vision through focused prototypes and working MVPs. Most Relevant customers include Claro, Banorte, T-Mobile, Telefónica, VHI, Sprint, AT&T, COX, VIVO Brazil, Hydro Quebec. During the execution of the project, all our customers consistently share a common goal: validating value early, reducing risk, and building momentum through tangible results.
Conversational Interfaces for Data Insights
Turn any business query into a visual, contextual answer in seconds… At Thingbook, we develop conversational insight engines that transform how non-technical teams interact with data. These aren’t simple chatbots, they’re AI-powered interfaces capable of answering domain-specific questions, generating SQL on the fly, retrieving relevant records, and visualizing the results through charts and narratives. Executives, marketers, and operations teams can explore business KPIs and trends by asking questions in natural language, receiving not just answers, but full context with embedded visuals and human-readable summaries. This dramatically reduces dependency on analytics teams, shortens decision cycles, and makes data accessible across the organization.
What’s included:
- Foundational Model & Cloud Agnostic.
- LLM-powered interfaces over SQL databases.
- Retrieval-Augmented Generation (RAG) pipelines with embedding stores.
- Auto-generated questions + charts (e.g., ECharts).
- Human-readable explanations + visual storytelling.
Example Use Cases:
- Marketing teams exploring customer behaviors without writing SQL.
- Operations teams querying KPIs and system health without technical support.
- Executives getting daily briefings from a conversational dashboard.
Our Customers:
Since the recent explosion in popularity of the available LLM foundational models, we have been working close with MATRIXX to deliver together a Text To Insight platform that allows Marketing Managers to discover the effect of AI in the conversion rates as well as define the ingredients that will make the next campaign just great. Currently we have several PoC ongoing to extrapolate this concept of Test to Insight to other verticals like smart factory and government.