CarbOptix Carboptix

Operate low‑carbon data centers with confidence

Carboptix unifies telemetry, invoices, grid signals, and work orders into one trusted platform. Get AI‑driven recommendations with citations, automate disclosures, and cut emissions without compromising uptime.

-35%
Infra Cost
+40%
Release Velocity
-50%
Incident MTTR
2.2×
Onboarding Conversion
Carboptix Portfolio Overview
Portfolio PUE
1.19
Grid Carbon Intensity
245 gCO₂e/kWh
Abatement Opportunities
  • Free‑cooling window + scheduled load shift
  • Fan curve tuning reduces 2% energy
  • PPA vs REC mix for Q4 disclosure
Tickets & Work Orders

Trusted by data‑driven infrastructure teams

The platform for measurable decarbonization

From ingestion to inference to audit‑ready reporting—all in one place.

AI-Driven Insights

LLM-powered RAG over your carbon + ops data to surface root causes, actions, and citations you can trust.

  • Hybrid retrieval (semantic + BM25)
  • Embeddings via OpenAI/HF/Cohere
  • On-upload & batch re-indexing

Unified Data Fabric

Ingest from meters, BMS, CMMS, utility APIs, invoices, and spreadsheets. Normalize into a governed data model.

  • Elasticsearch + pgvector document store
  • PDF/CSV ingestion & chunking
  • Schema evolution without drama

Enterprise-Grade Security

OIDC/JWT (RS256) with org-level row scoping, multi-tenant S3, presigned uploads, and audit trails.

  • JWKS signature validation
  • RBAC with role & org guards
  • Lineage + provenance on every write

Reporting that Persuades

Pixel-perfect branded PDFs, monthly schedulers, and traceable footnotes that hold up to scrutiny.

  • Celery beat scheduling
  • Citations linked to source docs
  • One-click exports (PDF/CSV)

Ontology + SHACL

BFO/CCO-aligned shapes for Factors, Bills, and Calculations. Automated validation at ingest & compute.

  • Cardinality & range rules
  • Issues table with provenance
  • Click-through lineage views

Cloud-Native Everywhere

Docker → K8s in minutes. Ingress with TLS (Let’s Encrypt). Horizontal scale without surprises.

  • Observability included
  • Autoscale-ready workloads
  • Zero-downtime deploys

Reference Architecture

A modern, observable, and secure stack designed for scale.

Ingestion Meters • BMS • Utility • PDFs Storage Elasticsearch • pgvector RAG & Ontology Hybrid retrieval • SHACL Reporting PDF • CSV • APIs Security & Governance OIDC/JWT • RBAC • Lineage • Audit • Ingress TLS

Simple, transparent pricing

Annual billing saves ~2 months.

Monthly Annual

Starter

For one site or pilot

8990 /yr
  • Up to 2 facilities
  • 10 users, SSO available
  • Dashboards & PDF exports
  • LLM Q&A with citations
  • Email support
Start Pilot

Growth

Scale across your portfolio

Most Popular
29990 /yr
  • Up to 25 facilities
  • Unlimited users & RBAC
  • Automated monthly reporting
  • Data lineage & governance
  • Priority support
Book Live Demo

Enterprise

Global + regulated environments

Custom
  • Unlimited facilities
  • VPC/VNet deployment
  • Custom models & integrations
  • SOC2/ISO27001 mappings
  • Named TAM & SLA
Talk to Sales

Resources

Explore the tech under the hood and see how customers deploy carboptix.

Get a tailored demo

Tell us about your portfolio and goals. We’ll tailor the walkthrough and ROI model.

  • Security review & architecture Q&A
  • ROI model with abatement forecast
  • Deployment options: SaaS, private cloud, on‑prem
By submitting, you agree to our privacy policy.

FAQs

How is carboptix deployed?

We support SaaS, private cloud, and on‑prem. Our reference path is containerized workloads with Helm charts for Kubernetes and Ingress + TLS via Let's Encrypt or your corporate CA.

What data sources do you support?

Native connectors for BMS/SCADA, utility APIs, smart meters, work orders, and invoice PDFs/CSVs. Anything else can flow through our HTTP or streaming ingestion adapters.

How do you handle security and compliance?

JWT/OIDC with JWKS validation, org‑scoped row‑level security, encrypted multi‑tenant object storage, audit logging, and mapped controls for SOC2/ISO27001. PII is minimized and access is least‑privilege.

What makes the RAG pipeline trustworthy?

Every answer includes citations linked to source documents. We combine semantic and lexical retrieval, chunk rigorously with metadata, and log provenance for each step.