commit dfae7ce876e6561622de930bdfdc11d48576d6bc
parent 73e0d377536791322c8ce5a998e68241f6b99dd3
Author: Silas Brack <57684859+silasbrack@users.noreply.github.com>
Date: Tue, 25 Feb 2025 21:57:51 +0100
Fixes (#1)
Diffstat:
6 files changed, 144 insertions(+), 34 deletions(-)
diff --git a/README.md b/README.md
@@ -1,2 +0,0 @@
-# My Blog
-
diff --git a/assets/cv.pdf b/assets/cv.pdf
Binary files differ.
diff --git a/assets/thesis.pdf b/assets/thesis.pdf
Binary files differ.
diff --git a/flake.lock b/flake.lock
@@ -0,0 +1,61 @@
+{
+ "nodes": {
+ "flake-utils": {
+ "inputs": {
+ "systems": "systems"
+ },
+ "locked": {
+ "lastModified": 1731533236,
+ "narHash": "sha256-l0KFg5HjrsfsO/JpG+r7fRrqm12kzFHyUHqHCVpMMbI=",
+ "owner": "numtide",
+ "repo": "flake-utils",
+ "rev": "11707dc2f618dd54ca8739b309ec4fc024de578b",
+ "type": "github"
+ },
+ "original": {
+ "owner": "numtide",
+ "repo": "flake-utils",
+ "type": "github"
+ }
+ },
+ "nixpkgs": {
+ "locked": {
+ "lastModified": 1736798957,
+ "narHash": "sha256-qwpCtZhSsSNQtK4xYGzMiyEDhkNzOCz/Vfu4oL2ETsQ=",
+ "owner": "NixOS",
+ "repo": "nixpkgs",
+ "rev": "9abb87b552b7f55ac8916b6fc9e5cb486656a2f3",
+ "type": "github"
+ },
+ "original": {
+ "owner": "NixOS",
+ "ref": "nixos-unstable",
+ "repo": "nixpkgs",
+ "type": "github"
+ }
+ },
+ "root": {
+ "inputs": {
+ "flake-utils": "flake-utils",
+ "nixpkgs": "nixpkgs"
+ }
+ },
+ "systems": {
+ "locked": {
+ "lastModified": 1681028828,
+ "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+ "owner": "nix-systems",
+ "repo": "default",
+ "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+ "type": "github"
+ },
+ "original": {
+ "owner": "nix-systems",
+ "repo": "default",
+ "type": "github"
+ }
+ }
+ },
+ "root": "root",
+ "version": 7
+}
diff --git a/flake.nix b/flake.nix
@@ -0,0 +1,29 @@
+{
+ description = "My gleam monorepo";
+ inputs = {
+ nixpkgs.url = "github:NixOS/nixpkgs/nixos-unstable";
+ flake-utils.url = "github:numtide/flake-utils";
+ };
+
+ outputs =
+ {
+ self,
+ nixpkgs,
+ flake-utils,
+ }:
+ (flake-utils.lib.eachDefaultSystem (
+ system:
+ let
+ pkgs = import nixpkgs {
+ inherit system;
+ };
+ in
+ {
+ devShells.default = pkgs.mkShell {
+ buildInputs = with pkgs; [
+ prettierd
+ ];
+ };
+ }
+ ));
+}
diff --git a/index.html b/index.html
@@ -4,58 +4,80 @@
<meta charset="utf-8" />
<title>Silas Brack</title>
<link rel="stylesheet" href="tufte.css" />
+ <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.3.0/css/all.min.css">
<meta name="viewport" content="width=device-width, initial-scale=1" />
</head>
<body>
<article>
<h1 id="tufte-css">Silas Brack</h1>
- <p class="subtitle">Machine Learning Engineer</p>
+ <p class="subtitle">Machine Learning Engineer —
+ <span title="Curriculum Vitae">
+ <a href="./assets/cv.pdf" style="background:none;"><i class="fa-solid fa-address-card"></i></a>
+ </span>
+
+ <span title="Github">
+ <a href="https://github.com/silasbrack" style="background:none;"><i class="fab fa-github"></i></a>
+ </span>
+
+ <span title="Linkedin">
+ <a href="https://www.linkedin.com/in/silasbrack/" style="background:none;"><i class="fab fa-linkedin"></i></a>
+ </span>
+ </p>
<section>
<p>
- I work on developing AI and ML models and deploying them to production, at Saxo Bank. This means both training
- in-house models and using pre-trained models (and combining the two). In my time at Saxo, I've worked on many
- projects, including both of the client-facing products my team has released:
+ I work on developing AI and ML models and deploying them to production at Saxo Bank. This means both training
+ in-house models and using open-source pre-trained models. In my time at Saxo, I've worked on many
+ projects, including multiple client-facing products my team has released:
+ </p>
+ <p>
+ <ul>
+ <li>
+ A representation learning model that learns which stocks <q>Others were interested in</q> for a given stock.
+ This model was our first to be integrated into the Saxo Bank platform and used by our customers. As of February
+ 2025, this model has seen a continuous increase in the number of interactions, and is currently sitting at
+ well over 100 000 monthly interactions.
+ </li>
+ <li>
+ A customer support chatbot which showed a 170% improvement over previous solutions in reducing client
+ interactions requiring human support. My implementation also reduced costs by around $150 000 a year,
+ eliminated several highly time-consuming manual steps, and was trivial to extend to all languages covered by
+ Saxo.
+ </li>
+ <li>
+ A recommender engine designed to be simple and extensible.
+ </li>
+ </ul>
</p>
- <ul>
- <li>
- A representation learning model that learns which stocks <q>Others were interested in</q> for a given stock.
- This model was our first to be integrated into the Saxo Bank platform and used by our customers. As of April
- 2024, this model has seen a continuous increase in the number of interactions, and is currently sitting at
- around 70 000 monthly interactions.
- </li>
- <li>
- A customer support chatbot which showed a 150% improvement over previous solutions in reducing client
- interactions requiring human support. My implementation also reduced costs by around $85 000 a year,
- eliminated several highly time-consuming manual steps, and was trivial to extend to all languages covered by
- Saxo.
- </li>
- </ul>
<p>Previously, I've also worked as a C# developer at MAN Energy Solutions.</p>
<p>
I did my master's in Mathematical Modelling and Computation at the Technical University of Denmark. During my
- thesis, I worked on implicit matrix algorithms for performing the Laplace approximation in Søren Hauberg's
- group in the section for Cognitive Systems of DTU Compute.
+ thesis<label for="sn-in-his-later-books" class="margin-toggle sidenote-number"></label>,
+ <span class="sidenote">
+ <a href="./assets/thesis.pdf">“Effortless Bayesian Deep Learning: Tapping Into the Potential of Modern Optimizers”</a>
+ </span>
+ I worked on implicit matrix algorithms for performing the Laplace approximation in Søren Hauberg's
+ group in the section for Cognitive Systems of DTU Compute. We also worked on applying the Laplace approximation to metric
+ learning, which we published at NeurIPS<label for="sn-in-his-later-books" class="margin-toggle sidenote-number"></label>.
+ <span class="sidenote">
+ <a href="https://arxiv.org/abs/2302.01332">“Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval,”
+ in <em>Advances in Neural Information Processing Systems</em>, 2023.</a>
+ </span>
</p>
<p>
Academically, I'm interested in machine learning and statistics, from deep learning to network science to
- Bayesian inference to forecasting. I love mathematics and would like to learn more, though I mostly care about
- its applications. I derive a lot of pleasure from sitting down and writing code, so I try to do as much of
- that as possible. Finally, I'm interested in the intersection of science and society, and science
+ Bayesian inference to forecasting. I love applied math and writing code, so I try to do as much of
+ those as possible. I'm also interested in the intersection of science and society, and science
communication is quite important to me. I spend a lot of time trying to learn more about how to effectively
communicate ideas and generally just enjoy teaching.
</p>
<p>
- In my spare time, I try to learn more about fields I'm less well-versed in—right now, that means
- economics, philosophy, paleontology, and history. Life is meaningless, so I just try to make the most of it by
- learning and teaching, appreciating beauty wherever I can, and by (mostly) doing things I enjoy. Because of
- this, I try to spend my time coding, reading, watching films and series, and listening to music. I also really
- like to cook.
- </p>
- <p>
- I'm on <a href="https://github.com/silasbrack">GitHub</a> and
- <a href="https://www.linkedin.com/in/silasbrack/">LinkedIn</a>.
+ In my spare time, I try to learn more about fields I'm less well-versed in. Right now, that means
+ economics, philosophy, paleontology, and history. I try to make the most of life by
+ learning and teaching, going on walks in nature, appreciating beauty wherever I can, and by (mostly) doing things I enjoy.
+ Because of this, I try to spend my time coding, reading, watching films and series, and listening to music.
+ I also really like to cook.
</p>
</section>
</article>
