Felix Hoffmann

DevOps | Computational Neuroscience | Reproducibility

[Felix Hoffmann @felix11h]

@menzenski Do you estimate and track refinement efforts in Jira? Not having a (good) workflow for this has been one of the pain points in our team.

[Felix Hoffmann @felix11h]

I get to play around with tools and platforms such as #Pulumi, #Django, #GitHubActions, #Heroku and #AWS. I'm very happy that I was able to join a team that allowed me to grow and learn so much in a short time. If you're interested in these things, let's chat!

[Felix Hoffmann @felix11h]

In an attempt to keep this profile at least somewhat current, a personal update: Last year I left my position in research and have since then been working as a #DevOps engineer at a fintech startup.

[Felix Hoffmann @felix11h]

@eLifeInnovation @lubianat @wikidata Is there a general document with all projects for the sprint? Is it publicly accessible? Thanks!

[Felix Hoffmann @felix11h]

From @tyrell_turing's lecture at #DLRLSS: "It has become clear that the original arguments against deep learning in the brain can be addressed and that there is a possibility for the brain to perform end-to-end optimization." (paraphrased)

[Felix Hoffmann @felix11h]

At #DLRLSS, Hugo Larochelle will start off the summer school with an introduction to deep learning. Looking forward to what's to come!

[Felix Hoffmann @felix11h]

In autumn last year I participated in a weekend hackathon hosted by d-fine, @dfine_hackathon. It was a fun experience! I now finally got around to putting together the code I wrote for our project. Have a look if you want to! https://t.co/XAA8cZNqlS

[Felix Hoffmann @felix11h]

@NotionHQ @cacarr @ProductHunt Very interested in Notion, seems to be exactly what I'm looking for. Only thing holding me back is the lack of Linux support. I wouldn't mind using the browser version, but offline support is essential to me – there's no positive update you can share in this direction?

[Felix Hoffmann @felix11h]

@RaymondRChua I am! I was wondering if there was a place to get to know people before the school. Are you going as well?

[Felix Hoffmann @felix11h]

I found this tutorial on Independent Component Analysis (ICA) very helpful. Explains different approaches to ICA and offers heuristic insight on why they work. The HTML document seems to be available since April 2000, which is nothing short of impressive. https://t.co/k5ZAWtCQkK

[Felix Hoffmann @felix11h]

@sharbat_c @polybiotique @ai_shwarya @scottrcole @SarkarSreyash @VenkRamaswamy @michaelerard @neuroecology @neurograce @goneparanoid @iampritishpatil @analog_ashley @martintelefont @lvlzay_sci Yes, I don't speak much French, I'm afraid!

[Felix Hoffmann @felix11h]

@sharbat_c @polybiotique @ai_shwarya @scottrcole @SarkarSreyash @VenkRamaswamy @michaelerard @neuroecology @neurograce @goneparanoid @iampritishpatil @analog_ashley @martintelefont Last week! @lvlzay_sci and I were reading Tintin. Each of us read different characters -- I'm somewhat proud of my Captain Haddock.

[Felix Hoffmann @felix11h]

What a venue to present my poster at #BernsteinConference! Stop by today at No. 77 to learn how anisotropy in spatial connectivity shapes the nonrandom structure in cortical circuits. Poster PDF: https://t.co/yKKUBoEs6a

[Felix Hoffmann @felix11h]

First experience giving a workshop talk at #BernsteinConference! Synaptic connections remodel at a high rate - how are nonrandom patterns in the network structure maintained? Possibly, through anisotropic axon morphology. Slides: https://t.co/u4loji3N0q

[Felix Hoffmann @felix11h]

@sharbat_c @scottrcole Looks interesting, thanks for sharing. Would love to discuss! We could have something like Slack (or open alternative) as platform for discussion?

[Felix Hoffmann @felix11h]

@sharbat_c Congratulations! You've put in a lot of effort for this! I'm looking forward to reading more about your project.

[Felix Hoffmann @felix11h]

@Profparker @cian_neuro Mouse auditory cortex. Reference: https://t.co/2axnPilef7

[Felix Hoffmann @felix11h]

In just 20 days more than half of cortical synapses are replaced by newly formed connections. Some few contacts seem to persist, resulting in power law distributed synaptic lifetimes. Can a simple stochastic model account for these synapse dynamics? -> Poster G.004 2pm! #FENS2018

[Felix Hoffmann @felix11h]

Digital version of poster is available on GitHub w/ LaTeX source: https://t.co/6bGIhxnZNQ #OpenScience Work supported by #SPP2041!

[Felix Hoffmann @felix11h]

@kateiyas Excellent, thanks for sharing this! I would love to see something similar for comp-neuro. One could think of a website where a list of reproducible computational neuroscience papers is maintained. Reproducible could mean a passing build/reproduction using a CI service...