~/About~/Systems~/Foundry~/Blueprint~/Journal~/Projects
Book a Call

I architect the systems
that run the business.

I build the backend systems that companies depend on but don't want to think about. Pricing engines, data pipelines, infrastructure that runs without someone watching it. I take responsibility for the system's outcome, not just its delivery.

How I Work

Infrastructure Design

I once built a pricing pipeline on a single Postgres instance. It worked until the catalog hit 4,000 products and the daily sync started timing out. Now I design for the data volume I'll have in a year, not the data I have today.

Autonomous Systems

The first version of my pricing engine ran silently for two weeks with a 30% miss rate. Nobody noticed because it never threw an error. Now every pipeline I build checks its own output and alerts when coverage drops.

Technical Strategy

Picking SQLite over Postgres when the data fits. Choosing batch over real-time when latency doesn't matter. Saying no to microservices when a monolith ships faster.

Crisis Management

A wallbox pricing bug slipped through to production because I skipped the data and went straight to the code. Three hours of debugging later, I found a single ordering issue in the pipeline. Now I always start with the data, not the logic.

Where I Work

Klevar is my practice.

This site is the proof layer. Klevar is the operating practice behind the work: custom automation architecture for pricing systems, supplier feeds, finance records, and workflow infrastructure.

Visit Klevar →

Client Feedback

He was experienced enough to explain that we had to take the project in a different direction than what I thought and he delivered.

Arno V.·Commentary Analyst, Forex Source

First class work...very quick and accurate.

Gerald D.

Prompt, easy communication, great output. Will hire again.

Emma P.

Selected Engagements

Inventory Allocation Simulator

Inventory Allocation Simulator is a planning console for distributors, retailers, and supply-chain operators that need to decide where scarce stock should sit before demand arrives. Teams need it when warehouse stock, SKU economics, lead times, demand history, and transfer lanes live in separate files or systems, and the planner has to defend a transfer before anyone knows which demand scenario will happen.

Case Study →

Returns & Claims Orchestration Engine

Returns & Claims Orchestration Engine is a logistics operations console for e-commerce teams that need to turn failed deliveries, returns, and carrier exceptions into owned claims cases. Teams need it when tracking tools say what happened but not who owns the next action.

Case Study →

Compliance Document Substrate

Klevar Docs is an internal document issuance and audit system for a multi-entity company. It gives company operators one API and CLI for creating invoices, credit notes, resolutions, letters, receipts, statements, and compliance documents while preserving the legal entity, number sequence, compliance artifacts, and proof trail behind each document.

Case Study →

Delivery Tracking Gateway

Delivery Tracking Gateway is a parcel tracking API for logistics and operations teams that need DHL, DPD, and GLS shipment updates normalized into one status stream. It is needed when staff or client systems have to check carrier portals one shipment at a time and cannot trust carrier-specific formats.

Case Study →

Dispute Resolution Workbench

Dispute Resolution Workbench is a finance exception queue for operators who need invoice mismatches, contract breaches, reconciliation failures, and webhook dead letters turned into owned cases. It is needed when every upstream system sees its own exception but nobody owns the full dispute through resolution.

Case Study →

SLA Penalty Settlement Engine

SLA Penalty Settlement Engine is a vendor-credit recovery system for procurement and finance teams that need service breaches turned into calculated credits, dispute records, and settlement artifacts. It is needed when contract clauses exist but evidence, math, and collection still live in spreadsheets.

Case Study →

Vendor Performance Intelligence Engine

Vendor Performance Intelligence Engine is a supplier risk monitoring system for procurement leaders who need vendor signals turned into current risk bands, frozen alert history, and review queues. It is needed when supplier issues become visible only after renewals, invoices, or missed obligations have already cost money.

Case Study →

Contract Lifecycle Engine

Contract Lifecycle Engine is a contract and obligation tracking API for legal-ops, procurement, and finance teams that need signed agreements turned into reviewed deadlines, renewal alerts, and audit-ready obligation history. It is needed when a company carries hundreds of vendor and customer contracts and one missed notice can turn into an unwanted renewal, missed SLA credit, or compliance gap.

Case Study →

Invoice Reconciliation Engine

Invoice Reconciliation Engine is an accounts payable control system for finance teams that need purchase orders, goods receipts, and vendor invoices matched before payment leaves the business. It is needed when invoice volume makes manual spreadsheet checks too slow, too inconsistent, and too easy to miss duplicate or inflated payments.

Case Study →

Sensor Telemetry Engine

Sensor Telemetry Engine is an IoT monitoring backend for operations teams that need high-volume sensor readings ingested, checked, and isolated by device before anomalies become missed incidents. It is needed when dashboards and basic alerts stop working because readings arrive faster than operators can inspect them.

Case Study →

Subscription Lifecycle Engine

Subscription Lifecycle Engine is a payment recovery and subscription-state service for SaaS operators that need Stripe failures turned into dunning, retries, notifications, and churn visibility. It is needed when failed charges sit in dashboards until revenue review instead of triggering recovery while the customer can still be saved.

Case Study →

Clinical Scheduling Engine

Clinical Scheduling Engine is a scheduling service for clinic operators that computes bookable appointment slots from live provider availability, room equipment, buffer times, and existing bookings. It is for clinics that need front-desk staff to offer valid appointments, prevent double-bookings, and refill cancellations before empty chairs become lost revenue.

Case Study →

Financial Compliance Ledger

Financial Compliance Ledger is an append-only audit record for finance and compliance teams that need payment discrepancies, escalations, and resolutions to stay defensible months later. It is for organizations where processor data, internal ledger data, and investigation notes must survive audit review without being rewritten by later edits.

Case Study →

Webhook Ingestion Engine

Webhook Ingestion Engine is a shared webhook intake service for engineering teams that need signed events received, stored, retried, and delivered without each product rebuilding its own handler. It is needed when Stripe, GitHub, Shopify, or internal webhooks are important enough that silent drops become customer-facing incidents.

Case Study →

Workflow Automation Engine

Workflow Automation Engine is a self-hosted workflow service for engineering teams that need webhook, cron, API, delay, and notification chains to run as trackable multi-step processes instead of inline code. A CTO, VP Engineering, or backend lead needs it when business workflows are spread across webhook handlers, scheduled jobs, and one-off scripts, and nobody can see which step failed, retried, or completed.

Case Study →

Event-Driven Notification Hub

Event-Driven Notification Hub is a shared delivery service for engineering teams that need application events to become email, Telegram, WebSocket, or SMS notifications without rebuilding the same notification code in every product. It is for teams with multiple backend services, tenant-specific templates, user preferences, deduplication rules, and delivery tracking that need one place to manage outbound messages.

Case Study →

Multi-Agent RAG Platform

Multi-Agent RAG Platform is an AI backend for document-heavy teams that need search and chat over their own files without losing source grounding, cost control, or auditability. It is for founders, CTOs, and engineering teams when a weekend "chat with your docs" prototype starts answering from the wrong sources, hallucinating details, or burning LLM budget without limits.

Case Study →

Transaction Reconciliation Engine

Transaction Reconciliation Engine is a backend for finance and accounting teams that need Stripe, PayPal, bank statement, and internal ledger records to agree before close. It ingests transactions from each source, normalizes them into one format, matches them through a four-rule confidence cascade, and turns unresolved rows into reviewable discrepancies.

Case Study →

Centralized Property Intelligence Hub

Centralized Property Intelligence Hub is a property data layer for product and AI teams that need one searchable record across listings, floorplans, image condition scores, and universal property IDs. It serves property platforms when frontend search or RAG answers need combined filters that no single service database can answer alone.

Case Study →

Idealo Price Optimization Platform

Idealo Price Optimization Platform is a marketplace pricing tool for e-commerce sellers that need live competitor positions translated into rank-aware price recommendations. It is needed when manual Idealo checks are too slow to protect buy-button rank, margin floors, and daily pricing decisions.

Case Study →

NBA Scenario Engine

NBA Scenario Engine is a player-prop scenario tool for bettors and analysts who need injury news translated into teammate projection changes before sportsbook lines move. It is needed when a ruled-out player changes minutes, usage, rebounds, and assists inside a short betting window.

Case Study →

Solar Pricing Engine

Solar Pricing Engine is a pricing pipeline for e-commerce operators that need supplier price lists, Shopify catalog data, competitor checks, and margin rules turned into daily price updates. It is needed when manual pricing work makes stale prices cost sales, margin, or both.

Case Study →

Frequently Asked

What kinds of problems do you solve?

Pricing engines that replace manual spreadsheet work. Data pipelines that pull from unreliable supplier APIs and normalize messy formats. Infrastructure that monitors itself and recovers without someone watching it.

How do you work with teams?

I work as an embedded technical partner or independent architect. For early-stage companies, I design the entire system. For established teams, I audit existing architecture and design the next iteration. I stay involved through deployment and production.

What industries do you specialize in?

E-commerce infrastructure, pricing systems, and data-heavy automation. Companies that process large volumes of product, pricing, or supplier data and need that processing to be reliable and fast.

Get Notified

New system breakdown? You'll know first.

Have a system that needs architecture?

Let's talk about what you're building.

Get in Touch