aselimc
- 60 skills
- 0 followers
- 1 week ago last updated
- ▌ Experiment Tracking · aselimcTrack ML experiments with W&B or MLflow. Log hyperparams, metrics, artifacts, and system metrics.
- ▌ Model Serving Infra · aselimcProduction model serving. Triton, TorchServe, BentoML. Dynamic batching, autoscaling, health checks.
- ▌ Sim To Real Transfer · aselimcBridge simulation to real hardware. Domain randomization, system ID, hardware abstraction.
- ▌ Architecture Diagrams · aselimcGenerate Mermaid diagrams for system architecture, data flows, class hierarchies, and API sequences.
- ▌ Hyperparameter Search · aselimcSystematic hyperparameter optimization with Optuna. Bayesian search with pruning.
- ▌ Quantization Pipeline · aselimcModel quantization for deployment. PTQ, QAT, INT8/FP16/INT4, hardware-specific calibration.
- ▌ Code Quality Standards · aselimcOrg-wide code quality standards for linting, formatting, type safety, and error handling.
- ▌ Infrastructure As Code · aselimcTerraform modules, state management, environment separation, cloud provisioning.
- ▌ Sensor Fusion Pipeline · aselimcMulti-sensor fusion. EKF/UKF state estimation, camera-LiDAR-IMU calibration, point cloud registration.
- ▌ Motion Planning Control · aselimcMotion planning and control. MoveIt2, Nav2, trajectory optimization, PID/MPC, ros2_control.
- ▌ Monitoring Observability · aselimcThree pillars of observability. Metrics (Prometheus), logs (structured), traces (OpenTelemetry).
- ▌ Pytorch Training Pipeline · aselimcStandard PyTorch training patterns with AMP, DDP, checkpointing, and best practices.
- ▌ High Performance Computing · aselimcPerformance optimization. Vectorization, memory layout, GPU acceleration, profiling, parallelism.
- ▌ World Model Implementation · aselimcLearned dynamics models for prediction and planning. RSSM, JEPA, Dreamer, imagination-based RL.
- ▌ Rl Algorithm Implementation · aselimcImplement RL algorithms (PPO, SAC, TD3) with proper numerical stability and vectorized envs.
- ▌ Self Supervised Pretraining · aselimcSelf-supervised visual representation learning. Contrastive, masked, distillation, and JEPA methods.
- ▌ Data Augmentation Strategies · aselimcDomain-appropriate data augmentation for CV, robotics, and NLP pipelines.
- ▌ Model Export And Optimization · aselimcExport PyTorch models to ONNX/TensorRT for production deployment. Validate numerical equivalence.
- ▌ Containerization Orchestration · aselimcDocker multi-stage builds, docker-compose, Kubernetes manifests, Helm, GPU support.
- ▌ Detection Segmentation Pipeline · aselimcObject detection and segmentation pipelines. COCO/VOC formats, DETR/YOLO/Mask2Former, mAP/mIoU evaluation.
- ▌ Frontend Component Architecture · aselimcFrontend component design. State management, routing, accessibility, responsive design, performance.
- ▌
- ▌ Ros2 Development · aselimcIdiomatic ROS2 development. Lifecycle nodes, QoS, launch files, TF2, custom messages.
- ▌
- ▌ Database Patterns · aselimcDatabase design. Schema normalization, indexing, migrations, query optimization. SQL and NoSQL.
- ▌ Numerical Methods · aselimcNumerical solvers. ODE/PDE, FEM, sparse linear algebra, stability analysis.
- ▌ Proposal Reviewer · aselimcReview technical project proposals for quality, overpromises, legal risk, and internal consistency. Use when the user asks to review, check, evaluate, or critique a project proposal, grant proposal, or any section thereof. Also use when asked to find overpromises, check for legally binding language, verify consistency between sections (objectives vs work packages vs deliverables vs budget), or assess proposal readiness for submission. Accepts proposals as docx, pdf, or plain text input.
- ▌ Pytorch Lightning · aselimc bundleDeep learning framework (PyTorch Lightning). Organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, implement data pipelines, callbacks, logging (W&B, TensorBoard), distributed training (DDP, FSDP, DeepSpeed), for scalable neural network training.
- ▌ Environment Design · aselimcRL environment design. Gymnasium API, vectorized envs, observation/action spaces, curriculum.
- ▌ Physics Simulation · aselimcPhysics engine patterns. Rigid body dynamics, collision detection, constraint solvers, determinism.
- ▌ Prompt Engineering · aselimcSystematic prompt design. Structured output, few-shot, chain-of-thought, and anti-patterns.
- ▌ Read Software Docs · aselimc bundleRead and navigate software package documentation websites. Use when the user provides a documentation URL and wants to understand an API, SDK, framework, or library. Triggers on requests like "read the docs at [url]", "look up [topic] in the [package] documentation", "what does [API] do according to the docs", or any task requiring fetching and comprehending online software documentation. Supports any HTML-based documentation site including Sphinx, Doxygen, ReadTheDocs, MkDocs, and custom doc sites. Common targets include NVIDIA Isaac Sim, Omniverse Kit, OpenUSD, ROS 2, PyTorch, and similar software packages.
- ▌ Rendering Pipeline · aselimcRendering for simulation and synthetic data. PBR, domain randomization, depth/mask rendering.
- ▌ Reward Engineering · aselimcDesign reward functions. Dense shaping, normalization, curriculum, and common pitfalls.
- ▌ Secrets Management · aselimcSecrets handling. Never hardcode. Vault, env vars, rotation, leak detection.
- ▌ Structured Logging · aselimcJSON-structured logging with consistent fields and correlation IDs across services.
- ▌ Citation Management · aselimc bundleComprehensive citation management for academic research. Search Google Scholar and PubMed for papers, extract accurate metadata, validate citations, and generate properly formatted BibTeX entries. This skill should be used when you need to find papers, verify citation information, convert DOIs to BibTeX, or ensure reference accuracy in scientific writing.
- ▌ Dependency Security · aselimcDependency auditing. CVE scanning, SBOM, lock files, container image scanning.
- ▌ PPTX · aselimc bundleUse this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.
- ▌ XLSX · aselimc bundleUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.
- ▌ API Design · aselimcREST and GraphQL API design. OpenAPI specs, error handling, pagination, versioning.
- ▌ Matplotlib · aselimc bundleLow-level plotting library for full customization. Use when you need fine-grained control over every plot element, creating novel plot types, or integrating with specific scientific workflows. Export to PNG/PDF/SVG for publication. For quick statistical plots use seaborn; for interactive plots use plotly; for publication-ready multi-panel figures with journal styling, use scientific-visualization.
- ▌ Ml Testing · aselimcML-specific testing. Shape validation, gradient flow, convergence sanity, data pipeline determinism.
- ▌ LLM Serving · aselimcDeploy LLMs for inference. vLLM, TGI, quantization for serving, OpenAI-compatible APIs.
- ▌ Git Workflow · aselimcStandard git workflow conventions for branching, commits, and PRs. Apply to all code changes.
- ▌ RAG Pipeline · aselimcRetrieval-Augmented Generation. Chunking, embedding, vector stores, retrieval, and generation.
- ▌ Scikit Learn · aselimc bundleMachine learning in Python with scikit-learn. Use when working with supervised learning (classification, regression), unsupervised learning (clustering, dimensionality reduction), model evaluation, hyperparameter tuning, preprocessing, or building ML pipelines. Provides comprehensive reference documentation for algorithms, preprocessing techniques, pipelines, and best practices.
- ▌ Test Pyramid · aselimcBalanced testing pyramid. Unit, integration, E2E tests. Coverage thresholds, fixture patterns.
- ▌ Transformers · aselimc bundleThis skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.
- ▌ 3d Perception · aselimc3D vision pipelines. Point clouds, depth estimation, 3D reconstruction, pose estimation, scene understanding.
- ▌ Auth Patterns · aselimcAuthentication and authorization. OAuth2, JWT, session management, RBAC/ABAC.
- ▌ Paper Writing · aselimcAcademic paper writing. LaTeX, conference templates, experiment tables, rebuttals, contribution positioning.
- ▌ CI CD Pipeline · aselimcCI/CD pipeline design. Lint, test, build, scan, deploy with environment promotion and rollbacks.
- ▌ LLM Finetuning · aselimcFine-tune LLMs with LoRA/QLoRA, instruction tuning, and alignment (DPO/ORPO).
- ▌ Edge Deployment · aselimcDeploy ML models on edge hardware. Runtime selection, containerized edge, OTA updates, MQTT telemetry.
- ▌ Proposal Writer · aselimc bundleWrite technical project proposals or specific sections of proposals for research grants, technology development, and innovation projects. Use when the user asks to write, draft, or create a project proposal, grant proposal, research proposal, technical proposal, or any section thereof (e.g., abstract, objectives, methodology, work packages, budget justification, impact statement, Gantt chart). Also triggers on requests mentioning proposal-related terms like 'deliverables', 'milestones', 'work plan', 'state of the art', or 'expected outcomes'. Supports EU Horizon, NSF, TUBITAK, ERC, and generic proposal formats. Outputs via docx or pdf skills. Does NOT handle market research reports or business plans.
- ▌ Threat Modeling · aselimcSTRIDE threat modeling. Attack surfaces, trust boundaries, severity ratings, ML-specific threats.
- ▌ Dataset Pipeline · aselimcPyTorch Dataset/DataLoader patterns for efficient, reproducible data loading and augmentation.
- ▌ PDF · aselimc bundleUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.
- ▌ DOCX · aselimc bundleUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of "Word doc", "word document", ".docx", or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a "report", "memo", "letter", "template", or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.