PyLossless
Python tooling for reproducible EEG preprocessing and quality-control workflows that researchers can audit and re-run.
View on GitHubHigh performance computing · Research software · Canada
Tyler Collins
Research Computing Consultant
I help researchers turn computational problems into reliable, scalable software and workflows. My work spans HPC systems, Python tooling, research analytics, EEG/MEG pipelines, performance profiling and optimization, and human-in-the-loop uses of LLMs for scientific teams.
- HPC
- Slurm
- Python
- Research software
- Research analytics
- GPU efficiency
- Coding agents
- LLM-assisted workflows
Latest post
What Does a GPU-Hour Actually Measure?
GPU-hours hide differences in GPU models, whole-device allocations, and MIG slices. Research reports need to say what was actually used and how the total was calculated.
Selected work
Highlighted projects
EEGStudyFlow
Workflow patterns for managing EEG studies from raw recordings through to analysis-ready outputs without re-deriving state.
View on GitHubWriting More Code with AI Agents
A practical talk about using AI-generated tests, checks, and diagnostic code without handing over the scientific decisions.
Watch the talkWriting
Recent posts
What Does a GPU-Hour Actually Measure?
You’ve probably seen a line like this in a paper or grant report: We used 10,000 GPU-hours. It looks like a useful answer. There’s a number and a …
Before You Rebuild Your RAG Pipeline, Try an Agent
If your RAG system keeps giving incomplete answers, try going fully agentic. I’m not saying retrieval-augmented generation is useless. It can make a …
Read the incident report before declaring AGI
I keep seeing people describe the OpenAI and Hugging Face security incident as evidence of AGI or how we are all doomed in the next year. The bots …
Your KPIs Aren't Telling You Enough
Is GitHub down if the website loads but GitHub Actions doesn’t work? Technically, no. A lot of GitHub is still available. You can browse, review …
Check what your AI plan does with your data
There is a lot of drama around OpenAI’s claimed solution to the Navier–Stokes existence and smoothness problem. I’m not going to try to evaluate the …
Talks & publications
Recorded talks
Writing More Code with AI Agents
Use AI slop to verify code you already trust
Serial Farms: Package Options and When to Switch to Farming
Submitting hundreds of small jobs without making a mess
Revisiting Cython: Is It Still Effective?
Five years after my first Cython webinar