I believe there are two questions to consider here:
First Consideration
It's important to consider whether a user can comprehend and navigate 80,000 projects. Generally speaking, I think not.
For example, Amazon.com may have 80,000 books "relevant" to my interests. But their interface provides me with filters and recommendations. I never need to confront 80,000 records to find just the few which interest me in the present moment.
Second Consideration
It's important to consider when to transmit those 80,000 records to the user interface (from server to client) and in what batch size should the client store and manipulate those records.
Prior to LTE mobile networks, I'd recommend that a small batch size (say between 10 and 100 records) be sent to the user interface. With SQL, it's possible to use the LIMIT keyword (as @mukul-kumar-jha suggested). You can also perform 'pagination' as described here (What is the best way to paginate results in SQL Server).
But, if those 80,000 records are not long (say, for example, less than 1 megabyte of data for the entire recordset), I can argue there may be benefit in transmitting those records to the client all at once then manipulating the data in the client with filters and such.
Whether to perform pagination at the database, or to filter within the client, depends a lot on a few factors: what's the users' connection speed; what do they pay for bandwidth; is the client sufficiently powerful to manipulate large data sets; how many round trips to the database need to occur for each user on average.
Good luck!