Sign in

Matt

@mattzcarey.com
629 followers 1.1K following 204 posts

ex pro windsurfer 🇲🇹 🌊 AI engineer. I work on streamlining access to large data sources, agents and retrieval - other interests in mech interp and data efficient fine-tuning.

PostsRepliesMedia
Matt @mattzcarey.com · 16/01/2025
Super fun to speak at a packed ai engineer london. My talk was about how we eval agentic RAG with Langsmith Thanks @robbiehudson.bsky.social and crew for having me and for organising a great event 🙏
080
Matt @mattzcarey.com · 13/01/2025
We have been building something for a while at StackOne with no idea what it should be called. Google just told us. We are building an extension an extension for the whole HR tech ecosystem.
030
Matt @mattzcarey.com · 13/01/2025
i've been building a thing 🦓 the api is coming together. the model compiled by zml I think this leans well on zml pros. One model, many backends, zero code change.
020
Matt @mattzcarey.com · 08/01/2025
so a bit like this
121
Matt @mattzcarey.com · 30/12/2024
London looking pretty this time of year
040
Matt @mattzcarey.com · 28/12/2024
made a lil website for my granny. She needed some photo storage for her (soon to be an online) bookshop. Build her a nice little admin panel and it has dark mode :)
040
Matt @mattzcarey.com · 27/12/2024
Good year. pushed pretty hard and got a lot better
040
Matt @mattzcarey.com · 27/12/2024
lfg
040
Matt @mattzcarey.com · 26/12/2024
Making this thing I have wanted for the last year :)
040
Matt @mattzcarey.com · 26/12/2024
python is so broken but the fact that this is a thing.. js ecosystem is no better wtf
120
Matt @mattzcarey.com · 21/12/2024
Check out the latency increase by using gpt-4o instead of a specific model tag This has to be a bug?
111
Matt @mattzcarey.com · 17/12/2024
I knew gemini was the best LLM as a judge. Deepmind just proved me correct. www.kaggle.com/facts-leader...
020
Matt @mattzcarey.com · 16/12/2024
meta^2 prompts ftw
010
Matt @mattzcarey.com · 12/12/2024
day 11 of Advent of ML (nice to be back) and we are talking scaling at test time. Do models reason? what is scaling at test time and will it lead to mythical AGI level reasoning? find out more about this new scaling law: mattzcarey.com/blog/advent-... #blogvent #adventofml
100
Matt @mattzcarey.com · 08/12/2024
I just completed "Guard Gallivant" - Day 6 - Advent of Code 2024 #AdventOfCode adventofcode.com/2024/day/6 brute forced this so hard. Not safe or fun. Enums in zig are cool tho
070
Matt @mattzcarey.com · 08/12/2024
I love when people think they know best online and end up arguing with the dude that built the actual thing. never gets old
020
Matt @mattzcarey.com · 06/12/2024
When the world wakes up to the power of evals. Tracing services gotta invest in their compute. 🙏 for the SREs
030
Matt @mattzcarey.com · 05/12/2024
Successful day. Turns out writing about best practises makes you more conscious of following your own advice. 60% still sucks but getting there.
020
Matt @mattzcarey.com · 05/12/2024
It's already day 5 of Advent of ML (also known as #blogvent) 😇 Finishing up on the mini dive into retrieval today. If you are not using these you probably should.. reranker! An upgrade to @cohere.com rerank v3.5 bumped our internal retrieval evals 5%!! mattzcarey.com/blog/advent-...
020
Matt @mattzcarey.com · 04/12/2024
I just completed "Ceres Search" - Day 4 - Advent of Code 2024 finally got to include a nice struct. Also how mental is that patterns typing. Strings actually being slices of unsigned integers never get old. #AdventOfCode adventofcode.com/2024/day/4
020
Matt @mattzcarey.com · 03/12/2024
we have bluesky enabled comments on my new latest Advent of ML blog post and forever more. you can just do stuff :)))))))) mattzcarey.com/blog/advent-...
020
Matt @mattzcarey.com · 03/12/2024
I just completed "Mull It Over" - Day 3 - Advent of Code 2024 Pretty grim but it works, zig feeling more natural Needle in a haystack indexOf is so cool and feels v natural. #AdventOfCode adventofcode.com/2024/day/3
130
Matt @mattzcarey.com · 03/12/2024
good days work. big increase on evals. where's my bonus?
240
Matt @mattzcarey.com · 02/12/2024
My ml engineer brain tells me this is just a tokenizer issue. My comp sci brain tells me the tokenizers are fundamentally the wrong abstraction and this is such a hack. Which is right?
210
Matt @mattzcarey.com · 02/12/2024
Day 2 - Advent of Code 2024 done still getting to grips with zig std lib. Been using ArrayLists over comptime + slices which I think is bad form - c'est la vie #AdventOfCode adventofcode.com/2024/day/2
040
Matt @mattzcarey.com · 02/12/2024
I'm way too excited to see the @pydantic.bsky.social crew cooking.
0123
Matt @mattzcarey.com · 01/12/2024
Day 1 done - "Historian Hysteria" - Advent of Code 2024 #AdventOfCode adventofcode.com/2024/day/1 look at those beautiful defers
010
Matt @mattzcarey.com · 01/12/2024
started advent of code in zig. wish me luck in related news the zig build system is awesome. zig build {day} runs the correct file on build :) code here: github.com/mattzcarey/a...
030
Matt @mattzcarey.com · 29/11/2024
rate limit a day keeps the doctor away
010
Matt @mattzcarey.com · 29/11/2024
just me?
030
Matt @mattzcarey.com · 27/11/2024
bad robot. you forgot your manners
020
Matt @mattzcarey.com · 25/11/2024
the AI tech equivalent of tupac vs biggie is happening rn. Lab brings out a paper saying structured output is bad and we should all stop using it. dot txt guys: skill issue -_-
130
Matt @mattzcarey.com · 20/11/2024
Yeeooww We broke 100 peeps listening to my drivel
020
Matt @mattzcarey.com · 20/11/2024
every ai/ml engineer
040
Matt @mattzcarey.com · 18/11/2024
We are gonna need a bigger boat! AI Demo Days @ Google Cloud Startup Hub - what a night! 🚀 5 demos, 100 attendees, 200+ waitlist
050
Matt @mattzcarey.com · 17/11/2024
I’d put money on this
000
Matt @mattzcarey.com · 15/11/2024
this just me?
010
Matt @mattzcarey.com · 13/11/2024
this guy doesn’t stop does he?
010
Matt @mattzcarey.com · 13/11/2024
and so on and so forth
020
Matt @mattzcarey.com · 13/11/2024
LOOK AT HIM. THAT’S MY QUANT. MY QUANTITATIVE. MY MATH SPECIALIST. NOTICE ANYTHING DIFFERENT ABOUT HIM? he's QWEN 2.5 CODER
020
Matt @mattzcarey.com · 12/11/2024
this plss thanks cutie
100