1,847
decisions/day
On the entire AXORI portfolio
+12%
margin
vs static strategy (fixed thresholds)
<3s
latency
Chat query response time
24/7
online
Never offline · always active
Atlas in 2 modes
Atlas' 2 modes
Atlas is a single brand · 2 different modes based on context. On the site it orients visitors. In the platform it orchestrates trading for customers.
Atlas Site Assistanthere on site | Atlas Trading Co-pilotin platform | |
|---|---|---|
| For whom | Site visitors · not logged in | Logged customers · brokers · funds |
| Knows | AXORI documentation · models · pricing · IT regulatory | Specific customer history · their assets · their performance · their choices |
| Memory | Current session only · zero persistence | 24-month decision history · trading patterns per customer |
| Does | Replies · orients · directs to demo CTA | Decides · executes trades · forecasts · proactive alerts |
| Real-time market | No · uses static knowledge 2024-2025 | Yes · hourly PUN · 48h weather · 7 EU markets · TERNA MSD |
| Example query | "What is MSD?" · "How much does Asset Management cost?" | "Customer C BESS: what are you doing tonight?" · "Show me Customer B P&L Q3" |
| Cost | Free · max 15 messages/h | Included in all plans · unlimited |
Want to try Atlas Site Assistant now? Click the button bottom-right (Atlas icon). Replies to any question about AXORI in real time.
A day in Atlas' life
What Atlas does in real time
Example of a standard operating day on the AXORI portfolio. Every decision is logged and auditable.
06:30
Refresh weather forecast + 24h PV/wind production prediction
08:45
Submit MGP offer for next day · 24 hour bands + market zone
11:00
MGP results · recompute strategy for intra-day MI
14:30
Decision: current PUN €142/MWh · sell extra 30% PV surplus over PPA
17:15
Customer C BESS: forecast Terna MSD call · hold 60% SoC for evening
21:00
MSD execution: sell 40 MWh secondary band at €198/MWh
23:50
Daily settlement · generate per-customer P&L · prep morning report
Data sources
What data Atlas uses
6 data sources combined in real time. No magic · just many inputs + a good decision model.
Hourly PUN (Day-Ahead Market)
7 EU markets (DE, FR, AT, CH, ES, GR, SI)
48h weather forecast (Open-Meteo + ECMWF)
12-month asset production history
Customer consumption patterns (if onsite PPA)
Active regulatory qualifications (UdD, MSD, Capacity)
Atlas vs static strategy
Why "fixed thresholds" aren't enough
Traditional management uses rules like "sell if PUN > €100/MWh". Atlas decides case by case.
❌ Static strategy (fixed rules)
Example
“If current PUN > €110/MWh → sell all PV surplus. Otherwise hold in battery.”
- ❌ Doesn't look at weather forecast (sell now but rain comes → better wait)
- ❌ Doesn't consider MSD opportunities (could sell €198/MWh in evening)
- ❌ Doesn't differentiate per asset (PV north ≠ PV south ≠ BESS ≠ wind)
- ❌ Doesn't learn from customer historical data
✅ Atlas (contextual decision)
Same case
“Current PUN €110/MWh, but forecast 14:00-17:00 raises to €142/MWh (heat wave + low national wind production). Also this BESS likely MSD-called at 19:00 (€185/MWh). Decision: sell 30% now, hold 70%.”
- ✅ Combines weather forecast + PUN prediction + MSD opportunities
- ✅ Decides per specific asset (zone, type, contract)
- ✅ Learns from per-customer historical patterns
- ✅ Result: +12% average margin vs fixed rules
Want to see Atlas in action?
Live dashboard demo, real examples on assets similar to yours, Q&A.
Request Atlas demo