v2.0 — Now with RAG Intelligence

Forecast Revenue
Before It Happens.

ChronoForge Pulse combines ensemble ML models with Retrieval-Augmented Generation to deliver sales forecasts your team can actually trust — grounded in real historical context.

95.2%
Forecast Accuracy
<200ms
Query Latency
4-Model
Ensemble Stack
Ensemble Forecast — Q4 2025 Live
+12.4%
vs Last Quarter
$2.4M
Predicted Revenue
0.82
Confidence Score
Integrates with your existing stack
SAP
Salesforce
PostgreSQL
Snowflake
Databricks
Power BI
Capabilities

Intelligence at Every Layer

From raw data ingestion to natural language insights — every piece of the forecasting puzzle, unified.

4-Model Ensemble

SARIMA, LSTM, XGBoost, and LightGBM work in concert — each model's strengths cover the others' weaknesses.

pgvector Search

HNSW-indexed cosine similarity over PostgreSQL. Sub-200ms retrieval across millions of scenarios.

Risk Intelligence

Automatic risk assessment — variance analysis, deviation factors, confidence scoring, and scenario modeling.

Multi-LLM Backend

Choose Claude, Mistral, OpenAI, or Gemini — or run fully offline with local embeddings.

Conversational Analytics

Multi-turn chat with context. Ask follow-ups, drill into regions — like a data analyst on speed dial.

Architecture

From Data to Decision in 4 Steps

Every query flows through a battle-tested RAG pipeline built for speed and accuracy.

01
Ingest & Embed

Sales scenarios are embedded via sentence-transformers or API providers and stored with rich metadata indexes.

02
Vector Retrieval

Your query hits the HNSW index. Top-k most similar scenarios are retrieved in milliseconds.

03
Context Assembly

Retrieved scenarios are filtered by region, category, and date, then formatted into a structured prompt.

04
LLM Generation

Context is sent to your chosen LLM. You get grounded analysis with cited sources — no hallucination.

Ensemble Engine

Four Models. One Prediction.

Each model brings a different perspective. The ensemble combines them into a more accurate forecast.

SARIMA
Statistical Time Series
91.8%
Accuracy
LSTM
Deep Learning
93.1%
Accuracy
XGBoost
Gradient Boosting
92.6%
Accuracy
LightGBM
Gradient Boosting
92.3%
Accuracy
Ensemble Accuracy: 95.2% ↑ 2.1% over best individual
Supply Chain Intelligence

Forecast Smarter. Plan Everything.

ChronoForge Pulse goes beyond prediction — it connects demand forecasts directly to production schedules, material plans, and procurement optimization.

MPS
Master Production Schedule

Converts demand forecasts into a period-by-period production plan. Balances capacity constraints against service level targets with chase or level strategies.

Available-to-Promise
Safety Stock
Capacity Planning
Demand Fencing
MRP
Material Requirements Planning

Explodes multi-level BOMs to calculate exactly what to order and when. Handles subassemblies, purchased parts, and raw materials with lead time offsetting.

BOM Explosion
Lead Time Offset
Net Requirements
Planned Orders
LOT
Lot Sizing Optimizer

Finds the optimal order quantities using five algorithms — from simple EOQ to Wagner-Whitin dynamic programming — minimizing total holding + ordering costs.

Wagner-Whitin
Silver-Meal
EOQ / POQ
Lot-for-Lot
Pipeline Output
// Ensemble Forecast → MPS → MRP → Lot Sizing { "total_cost": $2,847,350, "service_level": "94.2%", "planned_orders": 186, "bom_levels": 3, "best_lot_method": "wagner_whitin", "cost_savings": "12.4% vs lot-for-lot", "ai_analysis": "Claude Opus 4.5 insights…" }
Developer Experience

REST API. Zero Friction.

FastAPI-powered backend with auto-generated docs, typed schemas, and every endpoint you need.

POST/api/queryRAG query
POST/api/chatMulti-turn chat
POST/api/forecast/explainExplain prediction
POST/api/forecast/riskRisk assessment
POST/api/forecast/recommendAction items
GET/api/kb/statsKB statistics
POST/api/supply-chain/configureSet products & BOMs
POST/api/supply-chain/runMPS → MRP → Lot Sizing
POST/api/supply-chain/analyzePlan + AI insights
forecast_explain.sh
# Explain a forecast with historical context curl -X POST https://api.chronoforge.dev/api/forecast/explain \ -H "Content-Type: application/json" \ -d '{ "date": "2025-12-15", "sarima_prediction": 245000.00, "lstm_prediction": 251200.00, "xgb_prediction": 248800.00, "lgb_prediction": 247500.00, "ensemble_prediction": 248125.00, "confidence_interval_low": 232000.00, "confidence_interval_high":264000.00, "feature_importance": [ {"name": "month", "value": 0.34}, {"name": "promo", "value": 0.28} ] }'
Get Started

Interested in ChronoForge Pulse?

Our team will work with you to find the right plan for your organization. Please reach out and we'll get back to you within 24 hours.

Ready to See the Future?

Deploy ChronoForge Pulse in minutes. No infrastructure headaches. No data science PhD required.

Welcome Back

Sign in to your ChronoForge account

Please enter a valid email address
Password is required
Forgot password?
or
Full name is required
Please enter a valid work email
Password must be at least 8 characters
or

You're In!

Welcome to ChronoForge Pulse. Your AI forecasting agent is ready.

Data File Detected

How would you like to use this file?