You will get validated forecasts to support your business planning decisions
Project details
Unlike static reports that get outdated, I deliver production-ready Python pipelines for fresh predictions anytime. Complete with code, modern tools, data validation, and live walkthrough.
Single Product Forecast: Test forecasting on one product/metric before scaling. Includes pipeline, data validation, cross-validated model, and 1-hour walkthrough.
Business Forecasting: Multi-factor forecasting with up to 3 data sources. Models external variables like promotions, weather, or pricing. Ideal for inventory and revenue decisions.
Multi-level Forecasting: Generates forecasts across granularities (product, category, regional, totals) with consistency checks. Built for enterprises doing strategic planning across complex portfolios.
All tiers include data validation, stationarity checks, cross-validation, metrics, and documentation. No black boxes, just robust tools.
Data quality and requirements vary by project. These packages provide a starting point, but I'll craft a custom solution and transparent quote for your situation.
Message me for a custom quote!
Single Product Forecast: Test forecasting on one product/metric before scaling. Includes pipeline, data validation, cross-validated model, and 1-hour walkthrough.
Business Forecasting: Multi-factor forecasting with up to 3 data sources. Models external variables like promotions, weather, or pricing. Ideal for inventory and revenue decisions.
Multi-level Forecasting: Generates forecasts across granularities (product, category, regional, totals) with consistency checks. Built for enterprises doing strategic planning across complex portfolios.
All tiers include data validation, stationarity checks, cross-validation, metrics, and documentation. No black boxes, just robust tools.
Data quality and requirements vary by project. These packages provide a starting point, but I'll craft a custom solution and transparent quote for your situation.
Message me for a custom quote!
Machine Learning Tools
NumPy, pandas, PyMC, Python, PyTorch, scikit-learn, XGBoostWhat's included
| Service Tiers |
Starter
$350
|
Standard
$750
|
Advanced
$1,600
|
|---|---|---|---|
| Delivery Time | 3 days | 5 days | 8 days |
Number of Revisions | 1 | 1 | 1 |
Number of Model Variations | 1 | 1 | 1 |
Number of Scenarios | 1 | 1 | 1 |
Number of Graphs/Charts | 3 | 10 | 20 |
Model Validation/Testing | |||
Model Documentation | |||
Data Source Connectivity | - | - | - |
Source Code |
Frequently asked questions
About Tomislav
Decision Scientist | Data Scientist | Analytics Consultant
Amsterdam, Netherlands - 8:08 pm local time
CORE EXPERTISE
Machine Learning & AI: Python, PyTorch, Scikit-learn, XGBoost, Time Series Forecasting (Prophet, DARTS, Nixtla stack), Causal Inference, A/B Testing, Bayesian Modeling (PyMC)
LLM & RAG: OpenAI, HuggingFace, LangChain, LangGraph, ChromaDB, FAISS
MLOps & Cloud: Docker, Kubernetes, MLFlow, GitHub Actions, Databricks, AWS (S3, SageMaker, Redshift), Azure (ML, Synapse, ADF), databricks, model monitoring, drift detection
Data Engineering: PySpark, Pandas, Polars, ETL pipelines, data quality frameworks (Great Expectations)
SPECIALIZED SOLUTIONS
✓ Media Mix Modeling (MMM): Led Heineken's most successful analytics project—captured tens of millions in value first year. Bayesian attribution modeling, channel optimization, ROI measurement, budget allocation.
✓ IoT ML Solutions: Equipment monitoring, predictive maintenance, sensor data analytics, anomaly detection at scale—deployed solutions still running 3+ years later.
✓ RAG & LLM Applications: Production knowledge retrieval systems for financial services. Semantic search, document analysis, evaluation frameworks, prompt optimization.
✓ Entity Deduplication & Record Linkage: Fuzzy matching, probabilistic record matching at scale (millions of records), customer/product master data.
✓ Supply Chain & Logistics: Demand forecasting, inventory optimization, route planning, cost-benefit simulation.
WHY WORK WITH ME?
→ End-to-End Delivery: From data architecture to production deployment - I handle the full ML lifecycle so you don't need multiple specialists.
→ Business Impact: My solutions deliver measurable ROI and remain operational years after launch. One client brought me back to improve my original work - the ultimate validation.
→ Startup-Ready: Modern tech stack (FastAPI, Streamlit, containerized deployments), rapid prototyping, pragmatic solutions over academic theory.
→ Scientific Rigor: PhD in Physical Chemistry from University of Amsterdam - I bring systematic methodology to ambiguous business problems.
Looking for ML/AI expertise that delivers real business value? Let's discuss how my specialized skills can accelerate your data science initiatives.
Steps for completing your project
After purchasing the project, send requirements so Tomislav can start the project.
Delivery time starts when Tomislav receives requirements from you.
Tomislav works on your project following the steps below.
Revisions may occur after the delivery date.
1. Data Validation & Quality Assessment
Review all provided data sources for completeness, consistency, and quality issues. Check for missing values, duplicates, outliers, and temporal gaps. Document any concerns and confirm data structure matches requirements
2. Exploratory Analysis & Pattern Detection
Analyze historical patterns including trends, seasonality, and cycles. Identify time granularity characteristics and any structural breaks. Generate initial visualizations and summary statistics to understand data behavior.