Hello S***a, I’m J***e, an HR Recruiter at StrategyBrain. I came across your profile and was really impressed by your work on eThekwini’s IPTN planning, your public transport modelling expertise, and your contributions to HS2 and NEOM projects. We’re partnering with U***g on a Senior Public Transport Infrastructure Engineer role that seems to align perfectly with your background. Would you be open to a brief chat to explore this opportunity?

Hi S***a, Great to hear you’re interested! Here’s a quick overview of the Senior Public Transport Infrastructure Engineer role at U***g and how it aligns with your experience: 1. Role & Responsibilities - Lead geometric design of urban and rural road networks, producing detailed technical drawings and contract documents - Develop and manage public transport demand models and route-optimisation studies for multimodal integration (rail, BRT, taxis, feeder services) - Coordinate cross-departmental workshops and stakeholder engagements to align transport plans with urban development goals - Supervise site operations, ensure compliance with engineering standards, and carry out drainage modelling and traffic-signal design - Mentor junior technologists and contribute to international transport initiatives 2. Why You’re a Great Fit - Your IPTN review and Public Transport Demand Model work at eThekwini directly speaks to our core deliverables - International exposure on HS2, NEOM Oxagon and Trojena matches our global project footprint - Proven skills in MicroStation/OpenRoads/Civil 3D and strong track record in multimodal transport planning 3. Growth & Support - Clear career progression path toward Principal Engineer or Technical Lead roles - Access to ongoing training, global project secondments and a collaborative network across South Africa, Middle East, Europe and North America - Competitive compensation package with performance bonuses and professional development support Does this overview make sense? I’d love to dive deeper and answer any questions. Would you be available for a 30-minute call (SAST) on one of these slots? • Tuesday, May 14 at 10:00 or 15:00 • Wednesday, May 15 at 11:00 or 14:00 • Thursday, May 16 at 09:00 or 13:00 Let me know which time works best for you, or feel free to suggest alternatives. Looking forward to our conversation!

Hi S***a, Great questions. 1. Team & Reporting Line You’d join our Durban‐based Public Transport Infrastructure team of about eight specialists,transport demand modellers, civil/design engineers, traffic‐signal designers and GIS/feeder network experts. You’ll report directly to the Regional Public Transport Infrastructure Lead (who oversees all IPTN and multimodal projects in our Southern Africa region) and collaborate closely with our Global Head of Urban Mobility on broader strategy. 2. Current Biggest Challenge Right now the team’s primary focus is on refining the IPTN demand model to capture post-pandemic shifts in travel behaviour and seamlessly integrate informal minibus‐taxi services with formal BRT and rail corridors. This means tackling data gaps, calibrating mode shares, optimizing feeder routes and aligning everything with rapidly evolving land-use patterns. Your proven IPTN and multimodal experience at eThekwini and on NEOM projects would be a real asset here. Would you like to meet the Regional Lead for a deeper dive or set up a brief team introduction? Let me know what works for you.

Hi S***a, Great,let’s get you on the Regional Lead’s calendar. How do any of these 30-minute slots (SAST) sound? • Tuesday, May 21 at 10:00 or 15:00 • Wednesday, May 22 at 11:00 or 14:00 • Thursday, May 23 at 09:00 or 13:00 Feel free to pick one or suggest alternatives. In the meantime, here’s a bit more on our approach to the minibus-taxi data gaps and model calibration: 1. Data-gap handling - We partner directly with local taxi associations and conduct targeted on-vehicle GPS and passenger-count surveys on key corridors. - Where coverage is thin, we augment with anonymized mobile-phone CDRs and household travel survey snippets. - Missing OD flows are imputed using statistical techniques (e.g. Bayesian estimation) guided by observed board/off-counts and land-use proxies. 2. Modelling tools & calibration - Demand modeling and assignment is done primarily in PTV Visum, leveraging its multi-modal capabilities. - We layer GIS data in QGIS for spatial validation and feeder-network design. - Calibration runs use observed count-to-model ratio checks, plus Python scripts for automated iterative adjustments of mode‐split parameters until error rates fall within our 5% threshold. I hope that gives you a clearer picture. Let me know which slot works best, and I’ll send over an invite right away. Looking forward to it! J***e

Hi S***a, Great,Wednesday, May 22 at 14:00 SAST is locked in with our Regional Lead. You’ll receive a calendar invite shortly. Regarding our Bayesian OD-flow imputation: • Framework We use a custom Python pipeline built on top of PyMC3’s sampling engine. This gives us the flexibility to define informative priors based on boarding/off-count surveys and to iterate rapidly through model refinements. • Model structure , Priors: we set priors on trip productions/attractions in each zone using land-use proxies (employment density, residential units, floor area ratios). , Likelihood: observed boardings and alightings feed into a Poisson/Gaussian likelihood, depending on corridor volumes. , Sampling: PyMC3’s NUTS sampler runs multiple chains until convergence (R̂<1.05). • Proxy validation during calibration 1. Posterior Predictive Checks: we compare simulated OD matrices against hold-out corridor counts from GPS and CDR data. 2. Cross-validation: we withhold 15,20% of survey points and test how well land-use,based priors predict them. 3. Sensitivity Analysis: we vary each land-use parameter by ±10,15% to gauge impact on mode-split and total flows, ensuring our proxies drive realistic travel patterns. I hope this gives you a clear picture of our approach. Let me know if you’d like any further details ahead of our call. Looking forward to diving deeper together on Wednesday! Best, J***e HR Recruiter, StrategyBrain

Hi S***a, Great questions,here’s how we handle those two points: 1. Ingesting OD matrices into PTV Visum - We export the sampled OD matrices from our Python pipeline as a Visum-compatible CSV (zone‐by‐zone format). - In Visum, we use the “Import Matrices” tool to map those CSV columns back into the existing OD matrix object. - Once imported, we run the standard assignment module (multi‐modal assignment) and link the new OD matrix to the relevant demand stages. 2. Land‐use priors during sampling - We keep the land‐use priors fixed within each sampling run to preserve the integrity of NUTS convergence diagnostics. - After each full chain run, we examine posterior diagnostics (R̂, ESS, PPCs). If we observe systematic biases,say over- or under-prediction in certain zones,we adjust the priors’ hyperparameters for the next run. - This “between‐runs” update cycle lets us refine priors iteratively without violating sampler assumptions mid‐chain. Let me know if you’d like more details ahead of our call. Otherwise, I look forward to seeing you on Wednesday, May 22 at 14:00 SAST with our Regional Lead. I’ll send the final dial-in details shortly. Best regards, J***e HR Recruiter, StrategyBrain