user: danielcorin
413 karma | created 12 years ago
recent submissions
Stateful Agents and Basic Memory
(danielcorin.com)
2 pts|10 days ago|discuss
7 months ago|discuss
Gemini 2.5 Flash default time to first token increased from 2.0
(thoughteddies.com)
1 pts|7 months ago|discuss
1 pts|1 year ago|discuss
1 year ago|discuss
1 year ago|discuss
1 year ago|discuss
Cogno – Word Guessing Game
(cogno.fun)
1 pts|1 year ago|1 comment
Sonnet 3.5 can solve 40% of NYTimes Connections puzzles
(danielcorin.com)
2 pts|1 year ago|discuss
VLMs Hallucinate
(danielcorin.com)
2 pts|1 year ago|discuss
Protobuf-driven data extraction with language models
(github.com)
2 pts|1 year ago|discuss
VLMs Aren't Blind
(danielcorin.com)
2 pts|1 year ago|discuss
Evals: Unit Testing for Language Models
(danielcorin.com)
1 pts|1 year ago|discuss
A few options for LLM result streaming with SSE
(danielcorin.com)
1 pts|2 years ago|discuss
Simple sandboxed Python environment for LLM agent testing
(danielcorin.com)
3 pts|2 years ago|discuss
(Unsuccessfully) Fine-tuning GPT to play "Connections"
(danielcorin.com)
95 pts|2 years ago|50 comments
2 years ago|discuss
Figuring out how to use LLMs in production
(danielcorin.com)
2 pts|2 years ago|discuss
Shaping LLM Responses
(danielcorin.com)
3 pts|2 years ago|discuss
Using GPT-3.5 to Quickly Generate and Run Shell Commands
(danielcorin.com)
2 pts|2 years ago|discuss