Simon Novak

For a manufacturing industry client, we designed and implemented resilient business process automation using Temporal.io Durable Execution, with workers written in both .NET and PHP. The solution ensures that long-running, multi-step processes survive failures, retries, and infrastructure interruptions — a critical requirement in production environments where partial execution has real operational cost.

As part of this engagement, we also built a vector image generation service where incoming requests arrive via ASP.NET Core and are handed off to a Temporal worker responsible for producing precise SVG and DXF outputs. This decoupling keeps the API layer clean while offloading compute-intensive or stateful drawing logic to a durable worker.

Across two engagements, we delivered significant database performance improvements:

At a telecoms client, we diagnosed and resolved critically slow queries that were degrading daily operations — reducing execution time from 27 seconds to under 600 milliseconds through targeted SQL Server query tuning.

In a separate project, a 100× performance improvement was achieved through SQL Server query analysis and optimization, with the gains carried through to the application layer via Entity Framework — ensuring the improvements were durable and maintainable, not just raw SQL patches.

We designed and deployed a custom telephony solution for our own business number, which doubles as a reference implementation for clients with similar needs. The system handles inbound calls with language-aware routing: callers are offered the option to continue in English or leave a recorded message. Outside business hours and on weekends, calls route directly to voicemail. All interactions — caller numbers and any recordings — are logged directly into the CRM, enabling systematic follow-up without manual intervention.

Responsible developer for a core metadata platform managing streaming channels, sources, and broadcast data. Ongoing maintenance and development role; system operates stably in production.

Advising on ETL pipelines, data warehousing, and OLAP analytics for an organization employing people with disabilities. Currently implementing an automated basic costs calculation engine, integrating HRM data with employee counts and disability classification factors to drive rule-based cost allocation.

Implementing an external API integration to keep product and catalog data synchronized with daily automated updates. Project in progress.

During a consulting engagement with a Swiss energy technology company, I contributed across the full software development lifecycle — from feature development through to quality process improvement. Working within an agile team, I helped shape internal quality controls by introducing structured code review practices, refining branching strategies, and establishing clearer definition-of-done criteria that reduced defect leakage into production.

A key area of impact was advising the Lead Engineer on the adoption of Event Sourcing as an architectural pattern for their metering and invoicing domain. The system handled complex energy calculations for ZEV (Zusammenschluss zum Eigenverbrauch) topologies — a Swiss regulatory model that allows building communities or multi-tenant properties to share locally-produced energy (e.g. solar) and reduce grid costs collectively. In this context, the auditability of meter readings, consumption aggregations, and invoice calculations was business-critical: regulatory compliance required a complete, tamper-evident history of how every billing figure was derived.

I presented the case that Event Sourcing would address this directly — by persisting immutable domain events (MeterReadingRecorded, ConsumptionCalculated, ZEVShareApplied) rather than mutable state, the system would gain a natural audit trail without bolted-on logging. This also opened the door to temporal queries ("what did we calculate at time T?") and replay-based correction of historical errors — both highly relevant given the retroactive invoice adjustment requirements common in the Swiss energy sector.

The engagement reinforced my ability to bridge deep technical architecture decisions with domain-specific business requirements, and to communicate those trade-offs clearly to engineering leads and stakeholders operating under Swiss regulatory constraints.