Make better energy trading decisions.
Two ways to improve your energy purchasing and trading: Zorora, the AI intelligence platform purpose-built for African energy markets, and founder-led professional services from a team that has run solar portfolios and built data infrastructure at Google.
What you get
Part 1 · AI Intelligence
Zorora — Deep Research for Africa's Energy Markets
SAPP day-ahead prices. Eskom station build-up. NERSA tariff rulings. 80+ live market series across 6 providers. Zorora connects the sources your analysts already use and synthesizes them into cited research reports, acquisition diligence, and regulatory intelligence.
Deep Research
Cited Reports, Not Chat Answers
Ask Zorora a research question and get a structured report with inline citations, revenue projections, and embedded charts — not a chatbot response you can't trace back to a source.
Every claim cites its source. When sources disagree, Zorora tells you exactly where and why. The AI behind it — Nehanda v3 — is a 27B parameter model trained on 490+ energy policy documents to refuse fabrication, not optimize for plausibility.
- Structured reports with inline source citations
- Credibility-scored sources — high-confidence evidence ranks first
- Revenue estimates, tariff structures, and regulatory context
- Follow-up questions grounded in the same source material
Regulatory Monitoring
Track the Rulings That Move Markets
174 cached regulatory events across South Africa, Zimbabwe, and the US. NERSA tariff decisions, ZERA licensing notices, FERC orders, and state RPS compliance data — filtered by jurisdiction, fuel type, state, and year.
A NERSA notice about municipal tariff redetermination for Ekurhuleni, Msunduzi, or Madibeng changes the economics of every asset in those service areas. Zorora surfaces these events automatically — before your competitors find them in the gazette.
- 174 regulatory events cached with provenance tracking
- 3,264 US RPS compliance records across all states
- Filter by jurisdiction, state, fuel type, and year
- Configurable alerts when new rulings arrive
Asset Mapping
See Every Generation Asset on the Map
1,144 mineral deposits. 739 generation assets. Satellite imagery with night-light overlays. Click any asset and get its capacity, technology, ownership, and construction status — then launch a diligence search or research report directly from the map.
- Sentinel-2 satellite imagery with VIIRS night-lights
- Generation assets with capacity, operator, and status
- Mineral deposits and mining concessions
- One-click diligence from any asset popup
The Engine
Nehanda v3 — Purpose-Built for Energy RAG Synthesis
Zorora's synthesis engine is Nehanda v3 — a 27B multimodal model built on Qwen3.6 VL with native vision, 262K context, and epistemic alignment that carries across base model upgrades.
A Deep Research Model Trained on Epistemics
Nehanda v3 is trained on how to reason with evidence. Five stages of epistemic alignment teach the model to synthesize ranked sources with inline citations, preserve conflicts between sources instead of picking a winner, correct false premises instead of agreeing, and refuse to fabricate when the evidence isn't there.
The energy intelligence comes from the sources Zorora feeds it. The quality of the reasoning is Nehanda's job.
Built on Qwen3.6 VL with a 262K native context window and an integrated vision encoder, Nehanda v3 can process full regulatory documents in a single pass and ingest PDFs directly as images — no OCR pipeline, no artifacts.
Part 2 · Professional Services
When the problem needs judgment, not just software.
Zorora solves the repeatable problem. The hard problems — portfolio strategy, settlement architecture, regulatory positioning, diligence on a contested asset — need the people who built the software. Our team brings 14 years of Silicon Valley data infrastructure experience from Google, Looker, and CapitalG, on-the-ground solar asset management from Lumen Energy, and PV installation and SSEG operations from New Earth Energy. That combination — analytics rigor paired with Southern African operational reality — is the unfair advantage behind every engagement.
Portfolio Strategy
Forecasting and dispatch optimization strategy across multi-OEM portfolios. We bring the Lumen Energy asset-management playbook and the Google analytics toolkit to your generation fleet.
Settlement & Wheeling Architecture
With SAWEM launching, settlement infrastructure built on fragmented data will break. We design the data layer that makes wheeling, municipal reconciliation, and day-ahead settlement auditable.
Acquisition Diligence
Pre-acquisition technical and regulatory diligence on contested assets. Surface the non-financial risks — grid connection, NERSA rulings, counterparty exposure — that kill deals after the LOI.
Sibaya Casino, Durban — 2.31MW rooftop
The engineering manager noticed a 16% gap between current and previous municipality readings — the energy use simply didn’t add up. Asoba's intelligence platform diagnosed a R1.02M annual grid-bill leak, tracing 92.6% of the shortfall to chronic inverter-level hardware failures across 4 of 7 Huawei inverters.
By closing this “Savings Gap” with targeted maintenance, the site restores R624k in annual net profit. The JEPA anomaly detection model achieved 0.98 AUROC — near-perfect fault detection, weeks before traditional alerts would have fired.
Read the full case study →Start optimizing your energy decisions.
Run a live research query in Zorora, or book a strategy call with our founders to scope a portfolio, settlement, or diligence engagement.