# EPF Cohort 5 - Week 3
Date: 01/07/2024
### Updates
* Most of my research this week was primarily focussed on the networking layer; devp2p and libp2p. I found myself questioning the most fundamental of things like NAT categorizations (cones and restrictions and more, phew!), UDP multicasting and QoS levels, etc.
* Had a chat with a fellow (MaxDav) on discord on collborating on p2p-debugger as a side project.
* I also read some literature for optimizing finite field arithmetic on ARM for the Constantine library (issue: https://github.com/mratsim/constantine/issues/200). I got in touch with Agnish and Mamy from the team for discussing the scope of the issue and if I could pick it up as a project for EPF.
* [This](https://eprint.iacr.org/2021/185.pdf) paper particularly caught my eye because it has rather surprising and non-intuitive results.
* I tried out kurtosis for devnets and am truly amazed with the convenience of spinning up entire networks. It is super elegant.
* I also researched on the cryptography of verkle trees i.e. pedersen commitments and IPA. This was an interesting and unplanned undertaking.
I missed out on:
1. Reading the SSF paper mentioned in my previous updates.
2. Writing the article on swap-or-not shuffling.
3. Deep dive into peerDAS and sharding (except reading an article by a fellow on the topic)
4. Trying out consensus specs is still a floating todo
### Next Steps
* Progress forward with research on p2p networking layer. Follow up with MaxDav on the same.
* Try out constantine and if possible setup a template project for the ARM optimization.
* Look into PeerDAS and sharding
* Try out consensus specs, at the least
### Personal Notes
Through these weeks I have had very ambitious goals relative to what I have actually achieved. This gives me a clear measure of my pace and how much I can take on for the entire duration of Phase 3. This is great because planning will become easier.