Client stories
Names and sectors are used with permission. Details are trimmed to protect customer data.
We booked a debugging block after our catalogue API started returning stale prices every third request. Node Grove traced it to a shared Redis connection pool in cluster mode—not the pricing service we blamed. The write-up let our junior dev ship the fix Monday morning.
— Lerato N., back-end lead, retail marketplace (Pretoria)
The code review on our auth middleware was blunt in the right places. They flagged a refresh-token rotation bug we had deferred for two sprints. I wish the summary had included estimated effort per finding, but the severity ordering was enough.
— James K., CTO, health scheduling startup (Cape Town)
Our webhook retries were duplicating ledger entries. They sat with us through a three-hour screen share, rebuilt the failure in staging, and refused to sign off until we added idempotency keys. Tough session, exactly what we needed.
— Aisha R., engineering manager, payments integrator (Johannesburg)
Case: Stuck event consumers after a Node upgrade
A Durban logistics company upgraded from Node 16 to 20 over a long weekend. Monday morning, Kafka consumers lagged by hours. Internal staff suspected broker misconfiguration.
Node Grove compared heap profiles before and after upgrade, found a native addon still compiled for the older ABI, and paired with their vendor to rebuild the module. Consumers cleared within the same debugging block. The team later booked a review of their upgrade checklist.