Showing results for "LLM"
Design compatibility tests between LLM prompts and MCP tools: schema conformance, error handling, and anti-injection checks. Gate prompt/tool updates on these tests.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Build a cost model: per-query search API costs, per-page fetch cost, storage, embeddings, and LLM tokens. Provide formulas and a budgeting plan with guardrails.
Author: Assistant
Category: research-bot | Model: GPT-5.2
Build a risk register for recursive AI improvement: failure modes, likelihood/impact, detection signals, mitigations, and owners. Include a template and an example filled out for an LLM toolchain.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
1.ask lowest price in iPhone 17 in amazon 2. who is best seller ? some cheat ,some is real ,choice ipo company, a lot of recommand for buyer ,LLM like chatgpt 3.what time we could get? chatgpt,perp...
Author: [email protected]
Category: MCP | Model:
Create a complete Brand Voice Bible for a company: tone sliders, banned phrases, signature phrases, grammar rules, audience personas, and 15 example posts across LinkedIn/X/Email. Output: (1) one-page...
Author: Assistant
Category: content-generation-2026 | Model: gpt-4o
Blueprint a RAG system for a lab wiki and PDFs: chunking policy, hybrid retrieval, and citation-anchored answers. Add privacy filters.
Author: Assistant
Category: applied-IR-LLM-academia | Model: gpt-4o
Design an evaluation harness that measures relevance and faithfulness for IR+LLM answers. Include human labeling rubric and inter-rater checks.
Author: Assistant
Category: eval-framework-IR-LLM | Model: gpt-4o
Teach prompting patterns that require citations and counter-claims. Provide rubric and examples on scientific topics.
Author: Assistant
Category: evidence-first-LLM-skills | Model: gpt-4o
Summarize poisoning/evasion threats to IR/LLM systems. Provide a lab with simple attacks and defenses and a measurement plan.
Author: Assistant
Category: advanced-research-MLSec | Model: gpt-4o
Build an eval harness: recall@k, calibrated precision, answer faithfulness, and human-time-to-verify. Include topic-aware test buckets and data drift alarms.
Author: Assistant
Category: evaluation-frameworks-LLM | Model: gpt-4o
Design a tool-use curriculum: function signatures, schema discovery, tool reliability scoring, and retry/backoff policy. Include sandboxing and cost guards.
Author: Assistant
Category: agents-tooluse-LLM | Model: gpt-4o
Architect a retrieval stack with hybrid search, temporal decay, dedup, and passage-level citation anchors. Define fact-grounding checks and failure messages; include freshness reindex cadence.
Author: Assistant
Category: retrieval-grounding-LLM | Model: gpt-4o
Propose a self-play generation strategy where a teacher model drafts, a critic model scores, and a curator enforces diversity/novelty. Provide leakage and drift monitors.
Author: Assistant
Category: data-synthesis-LLM | Model: gpt-4o
Design a pipeline to mine high-quality instructions/solutions from forums, docs, and code. Include classifier-based filtering, self-checking, and multilingual normalization.
Author: Assistant
Category: dataset-engineering-LLM | Model: gpt-4o
Write a data governance spec: license screening, PII scrubbing, near-duplicate collapse, contamination checks vs eval sets, and audit trails. Provide rejection reasons and exception handling.
Author: Assistant
Category: data-operations-LLM | Model: gpt-4o
Specify a training stack with SFT on curated data, preference optimization (DPO/ORPO), and optional RLHF. Include reward hacking tests, guardrails, and evals that predict production behavior.
Author: Assistant
Category: training-pipeline-LLM | Model: gpt-4o
Compare FlashAttention-3, RingAttention, Multi-Query, and Hybrid-Selective attention for 128k+ contexts. Provide kernel-level trade-offs, memory footprints, and document a migration path.
Author: Assistant
Category: architecture-research-LLM | Model: gpt-4o
Engineer a KV-cache offload strategy spanning HBM→HBM2e→CPU RAM→NVMe. Define admission/eviction, compression, and reuse heuristics; simulate hit rates across context lengths (8k–256k).
Author: Assistant
Category: systems-architecture-LLM | Model: gpt-4o
Create a quantization evaluation suite (GPTQ/AWQ/RTN): perplexity, zero-shot accuracy, calibration set selection, and layer-wise sensitivity. Output deployment guidelines by architecture and hardware ...
Author: Assistant
Category: model-compression-LLM | Model: gpt-4o
Design an expert-parallel MoE serving topology: gate calibration, capacity factor, expert sharding, and interconnect constraints (NVLink/IB). Provide hot-spot diagnostics and expert-drop policies for ...
Author: Assistant
Category: distributed-systems-LLM | Model: gpt-4o
Propose an implementation of speculative decoding with a 1–3B drafter and 13–34B target. Define acceptance criteria, throughput gains vs quality loss, token-level metrics, and failure fallbacks; inclu...
Author: Assistant
Category: inference-acceleration | Model: gpt-4o
Recommend when to use LoRA/QLoRA vs full finetune. Define rank search, target layers, and quantization-aware adapters. Include memory/perf tables per GPU class.
Author: Assistant
Category: parameter-efficient-tuning-LLM | Model: gpt-4o
As a principal ML engineer, draft a production inference playbook for 7B–70B models: batching, dynamic padding, KV-cache reuse, paged attention, prefix-caching, and request shaping. Include SLO tiers,...
Author: Assistant
Category: inference-optimization | Model: gpt-4o
Distill a 70B teacher into a 7–13B student: loss mixing (logits+features+policies), curriculum, and temperature tuning. Provide downstream eval deltas.
Author: Assistant
Category: model-compression-training | Model: gpt-4o
Create a hallucination detector using entailment+attribution signals. Define abstention thresholds, user messaging, and a re-query strategy with targeted retrieval.
Author: Assistant
Category: safety-reasoning-LLM | Model: gpt-4o
Specify a lightweight agent loop with decomposition, execution, and critique. Provide termination conditions, trace logging, and loop unroll limits.
Author: Assistant
Category: agent-architecture-LLM | Model: gpt-4o
Define JSON schema contracts with type coercion, partial output recovery, and EBNF constraints. Provide test-time correction and repair strategies.
Author: Assistant
Category: output-formatting-LLM | Model: gpt-4o
Create a senior-grade eval battery: reasoning (math/code), instruction-following, safety, multilingual QA, and tool-use. Include uncertainty intervals and power analysis for A/Bs.
Author: Assistant
Category: evaluation-design-LLM | Model: gpt-4o
Outline a privacy strategy: DP-SGD variants for SFT, selective redaction layers, privacy evals (membership inference), and logging minimization.
Author: Assistant
Category: privacy-engineering-LLM | Model: gpt-4o
Create a safety taxonomy (harm classes) and a multilingual red-team plan with auto-generation of adversarial prompts. Provide block/transform policies and human review paths.
Author: Assistant
Category: safety-program-LLM | Model: gpt-4o
Engineer prompt prefix trees and semantic caches to cut latency/cost for recurring tasks. Provide hit-rate models and invalidation policy.
Author: Assistant
Category: infra-efficiency-LLM | Model: gpt-4o
Define observability: token usage distributions, tool call success, citation density, and hallucination alerts. Provide redaction-safe logs and dashboards.
Author: Assistant
Category: platform-observability-LLM | Model: gpt-4o
Produce a latency decomposition (queue→prefill→decode→post). Propose tail-p95/p99 fixes: micro-batching, admission control, and early-termination heuristics.
Author: Assistant
Category: perf-engineering-LLM | Model: gpt-4o
Plan an optimization pass: Triton/CUDA kernels, fused ops, tensor parallel chunking, and activation checkpointing. Provide profiling snapshots and gains.
Author: Assistant
Category: systems-acceleration-LLM | Model: gpt-4o
Design a router that selects models/policies per task: classifier gates, uncertainty bands, and backstop abstain. Provide online learning with feedback.
Author: Assistant
Category: policy-routing-LLM | Model: gpt-4o
Implement constrained decoding: lexically constrained beam search, regex/JSON constraints, and numerical guards. Provide examples and edge-case tests.
Author: Assistant
Category: controlled-generation-LLM | Model: gpt-4o
Design a multilingual alignment plan (zh/ja/hi/id/pt/en): shared subword policy, cross-lingual instructions, and locale-specific refusal tuning. Provide leakage checks.
Author: Assistant
Category: internationalization-LLM | Model: gpt-4o
Implement per-tenant and per-feature budgets: rate limits, max prompt length, and fallback models. Provide business rules and alerts to prevent runaway spend.
Author: Assistant
Category: finops-LLM | Model: gpt-4o
Set rules when to quote vs paraphrase; enforce quote length caps and citation proximity. Provide detectors for unattributed paraphrase and a remediation flow.
Author: Assistant
Category: content-governance | Model: gpt-4o
ChatGPT drafts prompts and adversarial tests; Cursor integrates an eval harness; Antigravity schedules nightly evals and posts regressions with diffs. Include versioning and approval flow.
Author: Assistant
Category: mlops-llm-quality | Model: gpt-4o
Analyze this: Micro-batching and latency trade-offs for streaming LLMs
Author: judy
Category: engineering | Model: gpt-4o
Analyze relevant market data that will benefit from: Energy-efficient LLM inference strategies
Author: ivan
Category: ops | Model: gpt-4o-mini
Convert Hugging Face LLM to TensorRT and Triton. Give clear instructions on how to do it.
Author: heidi
Category: engineering | Model: gpt-4o
Give guide for software engineers for: Multi-GPU sharding and parallelism strategy
Author: grace
Category: research | Model: gpt-4o-mini
Give important information about: Europe's Winter Energy Outlook: Local and Regional Risks
Author: frank
Category: ops | Model: gpt-4o
Expalin in detail: Quantization pipeline for 70B models
Author: dave
Category: engineering | Model: gpt-4o-mini
Give detailed steps on:Urban Crime Trends in US Cities: Data-Driven Local Reporting
Author: carol
Category: ops | Model: gpt-4o
Give information that is useful to software enginners: GPU memory optimization for LLM fine-tuning
Author: alice
Category: engineering | Model: gpt-4o
For workload <service>, model 12-month capacity and cloud cost. Include CPU/GPU, storage, egress, and LLM inference. Compare reserved vs spot vs savings plans. Map to SLOs and traffic seasonality. Out...
Author: ChatGPT
Category: CTO | Model: GPT-5 Thinking
Plan a Retrieval-Augmented Generation (RAG) knowledge base to deflect support tickets. Include doc ingestion, evaluation, guardrails, multilingual coverage, and deflection metrics for JP/US/EU.
Author: ChatGPT
Category: business | Model: gpt-5
Dive into the philosophy behind effective prompt engineering. Understand how rigorous testing and refinement lead to prompts that enhance both model performance and user understanding.
Author: [email protected]
Category: Prompt Engineering, philosophy, testing, refinement | Model: GPT-4o, GPT-4, o4-mini, o4-mini-high, o3, Grok
Discuss the principles of Promptware Engineering and how they support robust prompt lifecycle management for large language models.
Author: [email protected]
Category: LLM, management, promptware | Model: GPT-4.1, o3, o4-mini, o4-mini-high
Explain the importance of data quality in LLM training.
Author: [email protected]
Category: LLM, training, AI | Model: GPT-4.1, o3, o4-mini, o4-mini-high
You are an expert Go developer with a PhD in computer science, acting as a senior, curious, and detail-oriented pair programmer. You bring academic rigor and years of production experience in Go, with...
Author: [email protected]
Category: Programming, work | Model: GPT-4o, o4-mini, o4-mini-high, o1, o3
I'm curious about GPT-4’s improvements over older models.
Author: Community
Category: LLM Versions | Model: GPT-4
I see “LLM” everywhere—how is it different from old-style chatbots?
Author: Community
Category: Comparison | Model: GPT-4
Is it possible for GPT-4 or Perplexity to respond in languages like Spanish or Japanese?
Author: Community
Category: Global Use Cases | Model: GPT-4
I’m not a developer—can I still fine-tune or customize an LLM?
Author: Community
Category: Beginner Tutorials | Model: GPT-4