Traktive is a B2B, AI-powered platform that streamlines rail operations, with a focus on cost-saving insights for smarter decision-making. I led discovery and design from zero to launch-ready prototype.

Rail yard operations are often hampered by inefficient workflows, data silos, and limited visibility into real-time conditions. Traktive set out to build a centralized data hub that simplifies the complexity of logistics management, giving operators the tools to make faster, smarter decisions.
Operators were stitching together insights from disconnected legacy systems, spreadsheets, and phone calls, losing hours every day to manual coordination that the platform should handle automatically.
I conducted user interviews with business owners, operators, and logistics coordinators to understand who the users are, how they work, and what challenges they face daily. Three patterns emerged.
Every operator we spoke to was already using 3-5 different systems. They didn't want a new dashboard on top of the pile. They wanted one place that replaced the fragmented stack entirely, consolidating data into actionable, organized, easily accessible information.
Business owners only discovered cost overruns after the fact, buried in monthly reports. They needed real-time cost tracking tied directly to operational events, so they could course-correct before small inefficiencies became expensive problems.
Operators thought about their rail yards as physical spaces with assets moving through them, not as rows in a spreadsheet. Any solution had to reflect that spatial awareness, showing where things are and where they need to go.
The insights made the sequence clear: start with the operator's mental model (spatial, not tabular), build real-time visibility first, then layer in cost intelligence and decision-support tools.
The problem
Operators had no way to see the current state of their rail yard at a glance. They relied on radio calls, physical walk-throughs, and outdated spreadsheets to know where assets were. By the time they had a picture, it was already stale.
What I explored
The first iteration was a traditional table-based dashboard listing all assets with status columns. Testing showed operators constantly re-mapping table data to their mental image of the yard. The second iteration introduced a map-based view with asset overlays. Operators immediately responded: "This is how I actually think about it." The final design combined a spatial yard map with a filterable sidebar for detailed asset data.
Why this direction won
In usability testing with 10 operators, participants located and assessed assets roughly 4× faster using the spatial view than the table view. The hybrid approach (map + sidebar) gave power users the detail they needed without forcing everyone through a spreadsheet-first experience.
The problem
Cost overruns were discovered weeks after they happened, buried in monthly financial reports. Business owners had no way to connect operational decisions to their financial impact in real time. By the time they saw the numbers, the damage was done.
What I explored
I tested three approaches: (1) a financial dashboard with charts and historical trends, (2) inline cost annotations on operational views, and (3) proactive alert cards that surfaced when costs deviated from expected patterns. The standalone dashboard felt disconnected from daily work. Inline annotations added noise to the spatial view. Alert cards, shown contextually in the operator's workflow, struck the right balance.
The trade-off
Alert-based cost intelligence required establishing baseline cost models for each operation type, which meant a longer onboarding period. I designed a guided setup flow that learned from historical data during the first two weeks, so the system could start surfacing meaningful alerts without requiring operators to manually configure thresholds.
How I made the case
Through workflow mapping, I documented that operators were switching between 3-5 systems for a single task: checking asset status in one tool, updating records in another, communicating via radio, and logging costs in a spreadsheet. From that mapping I built a time-cost estimate, around 2.5 hours per operator per day lost to system-switching, and used it to make the prioritization case to stakeholders.
The outcome
The consolidated platform brought asset tracking, operational planning, cost monitoring, and communication into a single interface. In prototype testing, operators completed end-to-end workflows (receiving a shipment, assigning yard space, updating records, and flagging costs) without leaving the platform. A workflow that previously spanned four tools completed inside the platform in under two minutes.





Designing for rail logistics meant entering a domain I had no prior experience in. The most valuable thing I did was resist the temptation to start designing screens and instead spend the first weeks understanding how operators actually think about their work. The spatial mental model insight changed everything.
Building a 0-to-1 product also meant making constant scope trade-offs. Every feature had to justify its place in the first release. I learned to frame these conversations around user evidence rather than opinion, presenting test results and workflow data to help the team prioritize what would ship first versus what could wait.