- Published on
A synthesis of recent work on LLM forecasting agents, focusing on Bridgewater’s AIA Forecaster and why blending AI with market prices can beat either alone.
All Posts
All Posts
- bayesian-optimization (3)
- bleu-score (2)
- red-teaming (2)
- large-language-models (2)
- gaussian-processes (2)
- ai-benchmarking (1)
- machine-learning-engineering (1)
- kaggle-competitions (1)
- aide-framework (1)
- openai-research (1)
- ai-agents (1)
- causal-discovery (1)
- transformer-neural-network (1)
- attention-mechanism (1)
- layer-normalization (1)
- deep-learning (1)
- adia-lab (1)
- prompt-optimization (1)
- reverse-engineered-prompt-attack (1)
- large-language-model (1)
- transformers (1)
- weights-and-biases (1)
- natural-language-processing (1)
- google-colab (1)
- google-drive (1)
- python (1)
- json (1)
- generalized-product-of-experts (1)
- acquisition-function (1)
- occam-razor (1)
- hyperparameter-tuning (1)
- Published on
A survey of DoVer and related work on failure attribution, intervention-based debugging, and observability tooling for LLM agent systems.- Published on
A practical implementation-level walkthrough of how Ragas uses schema-constrained LLM judging (structured outputs + retries) and how key RAG metrics compute their final scores.- Published on
A satirical look at why skipping AI agent evaluations makes perfect sense if you don't value maintainability, customers, or long-term sanity.- Published on
How to model query quality, use bootstrapping, and report realistic RAG performance with RAGAS metrics and confidence intervals.