The library

quality-assurance · Designing QA workflows that balance automated gate checks with manual exploration when shipping must meet quality bars.

wordpress-pro · Extending WordPress with custom post types and REST endpoints when off-the-shelf plugins don't fit the domain model.

find-bug-root-cause · Isolating elusive bugs using binary search and hypothesis testing rather than random code inspection or blame.

retrospective-learning · Extracting actionable improvements from past work when structured reflection beats informal debrief discussions.

autoresearch · Multi-hypothesis research requiring tight iteration between rapid experiments and systematic discovery.

risk-management · Projects where risk blindsides kill delivery; proactive identification beats reactive firefighting.

implementing-llms-litgpt · Production-grade fine-tuning without abstraction layer overhead; ideal for understanding model internals.

quality-reviewing · Catching ecosystem mismatches before they reach production; complements automatic hook with external verification.

stakeholder-communication · Projects where stakeholder misalignment kills adoption; communication is connective tissue.

mamba-architecture · Long sequences (100K+ tokens), streaming inference, or memory-constrained deployments beating Transformer quadratic scaling.

deliverable-tracking · Client work where explicit tracking + dynamic labels beat unstructured task lists.

behavioral-neuroscience · Understanding brain constraints on behavior, design decisions informed by neural limits, or psychopharmacology reasoning.

nanogpt · Learning transformer internals, quick prototyping, or experimenting with variants without framework overhead.

freshdesk-triage · High-volume support queues where manual triage is the bottleneck.

clinical-foundations · Clinical practice where empirical assessment beats intuition for reliability and ethical practice.

rwkv-architecture · Combining RNN efficiency with Transformer quality when linear-time inference matters.

tables · Organizing structured data where tables provide clarity and queryability.

cognitive-psychology · Understanding decision-making biases, memory constraints, or attention bottlenecks in system design.

distributed-llm-pretraining-torchtitan · Large-scale pretraining where single-node limits are exceeded and distributed coordination is unavoidable.

developmental-psychology · Designing age-appropriate experiences, understanding delays, or intervening during critical periods.

huggingface-tokenizers · Fast, parallel tokenization when model-specific vocabularies and special tokens matter.

angle-generator · Marketing/comms where the same fact told through different angles reaches different segments.

research-methods-psych · Psychology research where internal/external validity and replicability are non-negotiable.

sentencepiece · Multilingual models or custom vocabularies where language-agnostic approach beats language-specific tokenizers.

nano-banana-blockrun · Generating production-ready images without API keys when you control on-chain payments via USDC micropayments, avoiding vendor lock-in.

social-psychology · Diagnosing why teams conform to poor decisions or explaining how organizational pressure shapes individual behavior beyond rational choice.

blameless-postmortem · Converting incidents into organization-wide learning while protecting the psychological safety required for honest incident reporting, especially in distributed systems and on-call rotations.

ecosystem-alignment · Keeping a custom Claude Code setup in sync with upstream platform updates, preventing drift and discovering new capabilities to adopt.

rca-causal-inference · Incidents with rich quantitative data where you need defensible, reproducible, mathematical causal claims (not just narrative RCA), especially when distinguishing multiple confounding factors.

peft-fine-tuning · Fine-tuning large models (70B+) on consumer hardware by training only 0.17% of parameters in a 6MB adapter, enabling cost-effective task-specific customization without full-model training.

rca-classical-methods · Linear, single-cause shop-floor incidents (5 Whys), or multi-pathway quality incidents (Fishbone), or safety-critical system design review (FTA/FMEA) where classical techniques have proven power.

rca-distributed-systems · Production incidents in microservice meshes (Kubernetes, Istio, Linkerd) where the fault could originate in any of 10-100 services and the causal path runs through network hops, retries, and timeouts.

nnsight-remote-interpretability · Running the same interpretability code on GPT-2 locally and Llama-405B remotely without code changes, enabling scalable mechanistic interpretability research on massive models.

rca-human-factors · Safety-critical industries (aviation, healthcare, emergency response, nuclear) where understanding why the operator's action was reasonable from their perspective reveals system design flaws to fix.

pyvene-interventions · guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework

distress-screening · screening for nonspecific psychological distress or tracking symptom burden over time

rca-systems-theoretic · investigating incidents in healthcare, aviation, nuclear, autonomous systems, distributed microservices, or any system where multiple act...

sparse-autoencoder-training · guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable f...

push · Solving push challenges

critical-reading · evaluating arguments, identifying bias, assessing source reliability, analyzing rhetoric, or synthesizing across multiple texts on the sa...

transformer-lens-interpretability · guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints a...

information-literacy · conducting research, evaluating sources, navigating digital information environments, teaching research skills, or addressing plagiarism ...

nemo-curator · GPU-accelerated data curation for LLM training

literary-analysis · interpreting literary texts, analyzing author craft, applying critical lenses, discussing theme and symbolism, or exploring how literary ...

ray-data · Batch inference and preprocessing on 100GB+ datasets across multi-node clusters.

riverpod-cancel · Flutter apps cancelling requests when user navigates or triggers rapid refreshes.

phonics-decoding · Teaching early literacy and diagnosing reading errors via Running Records.

grpo-rl-training · Teaching specific output formats (XML, JSON) and verifiable tasks without preference pairs.

reading-comprehension · Teaching comprehension when decoding is automatic and meaning-construction needs scaffolding.

miles-rl-training · Training 1TB+ MoE models with speculative RL for 25%+ rollout speedup.