Results for “hzl”

Tracks multi-step work across sessions using the hzl CLI, coordinating sub-agents with projects, tasks, checkpoints, and leases.

Post, claim, and settle agent tasks on the AZZLE protocol (Base mainnet, USDC escrow). Requires Bankr for swaps, approvals, and on-chain execution.

Quantize large language models to 8/4/3/2/1-bit precision without calibration data, using multiple optimized backends for deployment with vLLM or HuggingFace Transformers.

Train or fine-tune language and vision models using TRL or Unsloth on Hugging Face Jobs cloud infrastructure, with support for SFT, DPO, GRPO, and reward modeling, plus GGUF conversion for local deployment.

Expert knowledge for Azure Cloud Hsm development including troubleshooting, best practices, limits & quotas, security, and integrations & coding patterns. Use when configuring auth, key lifecycle, PKCS#11 cert storage, Event Hubs logging, or checking HSM limits, and other Azure Cloud Hsm related development tasks. Not for Azure Dedicated HSM (use azure-dedicated-hsm), Azure Payment Hsm (use azure-payment-hsm), Azure Key Vault (use azure-key-vault), Azure Attestation (use azure-attestation).

Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

贺炜（体育解说）认知与表达框架（压缩蒸馏）：文学比喻嵌入赛况、克制激情、终场金句… 触发：足球诗人解说 等。非煽动球迷对立

Quantize large language models to 8/4/3/2/1-bit precision without calibration data, using multiple optimized backends and integrations with HuggingFace Transformers, vLLM, and PEFT/LoRA.

Operates Hetzner Cloud resources via the hcloud CLI, covering server, image, network, and load balancer management with read-only and write operations.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

Search, filter, and browse high-signal academic papers using the Paperzilla CLI.

Assess SSL/TLS server configurations using the sslyze Python library to evaluate cipher suites, certificate chains, protocol versions, HSTS headers, and known vulnerabilities like Heartbleed and ROBOT.

Write, read, and validate HUML documents, converting between YAML/JSON/TOML and HUML for human-readable configuration files.

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag) using standardized prompts and metrics. Supports HuggingFace, vLLM, and API backends.

Expert knowledge for Azure Payment Hsm development including troubleshooting, best practices, decision making, architecture & design patterns, security, and configuration. Use when configuring Payment HSM VNets/FastPath, payShield Manager access, HA/DR topologies, SKUs, or traffic inspection, and other Azure Payment Hsm related development tasks. Not for Azure Dedicated HSM (use azure-dedicated-hsm), Azure Cloud Hsm (use azure-cloud-hsm), Azure Key Vault (use azure-key-vault), Azure Information Protection (use azure-information-protection).

Quantize LLMs to 8/4/3/2/1-bit precision without calibration data, using multiple backends and HuggingFace/vLLM integration.

Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.

Runs a nightly automated retrospective on autonomous trading strategy performance, computing win rates, fee drag, holding period buckets, direction bias, and producing data-driven improvement suggestions.

Write and optimize Kusto Query Language (KQL) for Azure Data Explorer, Log Analytics, and monitoring workflows. Use this skill when users ask for KQL queries, telemetry analysis, incident investigation, or query optimization.

Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware, with backend selection between vLLM, Transformers, and accelerate.

Manage cryptographic keys in Azure Key Vault and Managed HSM using the .NET SDK. Create, rotate, encrypt, decrypt, sign, and verify keys with KeyClient and CryptographyClient.

Resolves, generates, and operates on media assets (audio, images, icons, logos, voice, color grades, LUTs) for HyperFrames projects, using a local cache and the HeyGen CLI for free-usage catalog search and TTS.

Assesses and automates upgrades of Azure workloads between plans, tiers, or SKUs, and modernizes Azure SDK for Java dependencies in source code.

Manage Azure PostgreSQL Flexible Server deployments using the .NET SDK. Create servers, databases, firewall rules, configurations, backups, and high availability.

Manage Azure MySQL Flexible Server deployments using the .NET SDK. Create servers, databases, firewall rules, configurations, backups, and configure high availability.

Routes Azure VM and Virtual Machine Scale Set (VMSS) requests to the right workflow for recommendation, creation, troubleshooting, capacity reservation, and essential machine management.

Assess and improve the reliability posture of Azure Functions and App Service applications by scanning for zone redundancy, ZRS storage, health probes, and multi-region failover, then driving staged remediation with user confirmation.

Hugging Face Hub CLI (hf) — search, download, and upload models and datasets, manage repos, query datasets with SQL, deploy inference endpoints, manage Spaces and buckets.

Fetches real-time London transit data from the TfL API, including Tube and bus arrivals, line status, disruptions, journey planning, and route information.

Expert knowledge for Azure Web PubSub development including troubleshooting, best practices, decision making, architecture & design patterns, limits & quotas, security, configuration, integrations & coding patterns, and deployment. Use when using WebSockets/MQTT, Socket.IO, Functions bindings, geo-replication, or Premium autoscale in Web PubSub, and other Azure Web PubSub related development tasks. Not for Azure SignalR Service (use azure-signalr-service), Azure Event Hubs (use azure-event-hubs), Azure Service Bus (use azure-service-bus), Azure Relay (use azure-relay).

Audit Azure Blob and ADLS storage accounts for public access exposure, weak or long-lived SAS tokens, missing encryption at rest, disabled HTTPS-only traffic, and outdated TLS versions using the azure-mgmt-storage Python SDK.

Build and run custom fuzzers with LibAFL, a modular Rust fuzzing library, including drop-in libFuzzer replacement and custom component configuration.

Merlin Chain: largest Bitcoin L2 by TVL (2026), ZK rollup with EVM, 150+ dApps. Federation peg with planned trust-minimization. USE WHEN: evaluating Merlin for deployment, BTC DeFi.