Context and Challenges:
- Aggressive growth leading to a +50M$ fundraising
- As a side effect, the system was experiencing serious scalability issues:
- Frustrating load times for end users on key pages, for example the scheduling page
- Timeouts and 500 errors were popping up randomly
- The overall performance of the system was starting to have a negative impact on user retention, hence on revenue.
- Little DevOps capabilities which created an opportunity to upskill the tech team
Duration: 4 weeks
Solutions and Impact:
- Divided by a factor of 10 the response time, hence significantly increasing the quality of service of the platform by prioritizing the most impactful / easiest-to-implement actions:
- Algorithmic optimisations
- N+1 request optimisations
- Index configuration on the database
- Runtime reconfiguration and optimisation
- Infrastructure optimization
- Co-creating with the client's team the methodology and management rituals to help solve the remaining slow requests that required deep structural changes
- Transfer to the client's team the ability to continuously monitor performance and to intervene quickly when needed to stay on target