Analytics scripts
Engineers author and reuse Python analysis logic with in-platform editing and explicit bindings to authorized vehicle signals.
Back to Work
FEV Software GmbHGermany · RemoteSep 2025 – Jan 2026 · ContractNDA · Internal product
FLEDEM
Fleet-analytics platform for FEV Software GmbH
Contracted as a full-stack engineer inside FEV's fleet analytics platform. Worked on core analysis workflows, data configuration flows, and engineering-facing tools — across a large React + ASP.NET Core + MongoDB codebase processing CAN signal data from connected vehicle fleets.
ASP.NET CoreReactTypeScriptMongoDBPythonSignalRApache ParquetAzure Blob StorageTailwind CSS
NDA-safe system portrait
Vehicle telemetry, analysis pipelines, and fleet-scale signal processing—visualized without exposing private product screens.
4
core modules contributed
Cross-stack
frontend + backend + infra
4 months
Sep 2025 – Jan 2026
Enterprise
scale & complexity
Overview
Making complex fleet data feel usable
FLEDEM is FEV's internal fleet-analytics platform — used by test engineers to define, validate, and reuse analysis logic against CAN-signal telemetry from connected vehicle fleets. Before this engagement, configuration-heavy flows were fragmented and engineers leaned on external scripts to get things done.
My contract focus was reducing that friction inside the product. I built features that let users define and reuse analysis logic directly inside the platform, simplified complex create/edit flows, and standardized UI patterns across configuration-heavy screens so the platform felt consistent end-to-end.
Beyond shipping features, the engagement included a meaningful amount of platform-quality work: contributing to automated testing, API documentation, security review, and build-quality improvements across the large frontend/backend codebase, plus role-based access patterns, reusable notification/confirmation systems, and design-system alignment for future modules.
Analysis workflow
Configuration becomes a reusable engineering system.
FLEDEM brings the pieces of fleet analysis—scripts, calibrations, signal mappings, and detected events—into one controlled product workflow instead of leaving them in disconnected engineering tools.
Project configuration
Calibration files connect to projects, scripts, and analysis packages through versionable records and managed attachments.
Signal mapping
CAN channels, units, logger slots, and stable configuration identifiers make analysis inputs explicit across revisions.
Detected events
Event definitions, severity, evidence, filters, and statistics turn analysis output into a workflow users can investigate.
Engineering Challenges
The hard parts I worked on
Configuration-heavy flows that confused users
Solution · Refactored dominant create/edit patterns into a consistent stepper + side-panel model, deduped form components, and aligned validation feedback so the same edit primitives behaved identically across modules.
Outcome · Configuration-heavy screens now share a clearer interaction pattern instead of teaching a different editing model in each module.
Bringing analysis logic inside the platform
Solution · Built the in-platform Scripts module — Monaco-editor authoring, channel-mapping UI for binding script inputs to CAN signals with unit conversion, and persistence so scripts could be reused across calibrations and analysis packages.
Outcome · Analysis logic moved from one-off external scripts into a versioned, reusable, fleet-scoped resource inside the product.
RBAC across modules without repeating yourself
Solution · Contributed to role-based access patterns and reusable confirmation/notification components so new modules inherit the same RBAC + UX guardrails by default rather than reimplementing them.
Outcome · Future modules align with the same access-control + design-system patterns from day one.
Quality bar across a large codebase
Solution · Beyond features, contributed to automated testing, API documentation, security review, and build-quality improvements across the frontend/backend codebase — especially around the modules I touched.
Outcome · Modules I owned shipped with the test, docs, and security review needed for handover.
Enterprise architecture
A cross-stack path from vehicle signal to product decision.
The platform joins a typed React product surface, .NET domain services, real-time events, analytics execution, and multiple storage formats behind consistent access and interaction patterns.
React engineering tools
TypeScript, Monaco, charts, and shared interaction patterns support dense configuration work without hiding domain detail.
ASP.NET Core services
REST contracts, authentication, permissions, and module logic provide stable boundaries for cross-team delivery.
Python and SignalR
Analytics execution and real-time event delivery connect vehicle-derived results back to the product experience.
MongoDB and Parquet
Operational documents, binary identifiers, time-series output, and file assets use storage suited to their access patterns.
What I Delivered
Outcomes from a 4-month contract
- 01
Reduced workflow friction for fleet engineers by enabling them to define, validate, and reuse analysis logic directly inside the platform — instead of relying on fragmented external scripts.
- 02
Improved platform usability and consistency by simplifying complex create/edit flows, standardizing UI patterns, and making configuration-heavy screens easier to operate.
- 03
Strengthened product reliability by contributing to automated testing, API documentation, security review, and build-quality improvements across a large frontend/backend codebase.
- 04
Supported enterprise readiness by contributing to role-based access patterns, reusable notification/confirmation systems, and design-system alignment for future modules.
- 05
Operated as an independent contractor within a distributed engineering team, owning end-to-end delivery of assigned features.
FLEDEM is FEV Software GmbH's internal product. Screenshots and proprietary architectural details are excluded under NDA; the contributions described above are reflected in the engineering record and can be verified with the FEV team on request.
Available for senior contract work
Need an engineer for your enterprise platform?
I join established teams to ship high-stakes product work across React, .NET, MongoDB, Python, and SignalR.