Infinite Logscape Dashboard

Inspiration

Manually going through log data and understanding it, is a time consuming process. Additionally, finding specific information, especially if it is not explicitly match a keyword, is particularly difficult.

What it does

We are introducing a solution that provides statistical insights into log data. This includes looking at the frequency of logs during specific time periods and the type. It also includes a semantic search feature - allowing users to find what they want without having to know the specific words or phrases.

How we built it

We built the frontend using Flutter, the backend using Django and all other functionality in Python. Our main feature is using embeddings, generated using a model from the SentenceTransformer library.

Challenges we ran into

Our main challenge was difference in the skillset as only one of the team members was proficient with Flutter, it became the bottleneck of the development. Also, we iterated on ideas, abandoning them, resulting in precious time spent on the feutures, not added to the final product.

Accomplishments that we're proud of

We would say that embeddings was quite an advanced feuture, which greatly simplifies the search, if the user does not remember the exact wording of a particular log message. We believe such idea will can have many real-world applications.

What we learned

We learnt about the teammwork and the importance of making sure that strengths of each team member are utilised. It allowed use to get decent UI along statistical insigths and efficient utilisation of machine learning.