{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","slug":"ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","name":"AI-Sys-Sp22 Machine Learning Systems - University of California, Berkeley","type":"product","url":"https://ucbrise.github.io/cs294-ai-sys-sp22/","page_url":"https://unfragile.ai/ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","categories":["productivity"],"tags":[],"pricing":{"model":"unknown","free":false,"starting_price":null},"status":"inactive","verified":false},"capabilities":[{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_0","uri":"capability://planning.reasoning.structured.peer.review.facilitation.for.research.papers","name":"structured-peer-review-facilitation-for-research-papers","description":"Organizes graduate students into rotating reviewer roles (summarizer, systems expert, applications expert) for weekly research paper evaluations, with written reviews submitted before synchronous discussion sessions. The course implements a mini-program-committee (mini-PC) meeting format where students present papers in different roles across multiple weeks, simulating academic conference review workflows. Reviews are collected asynchronously via email/Slack, then discussed in real-time Monday seminars with instructor feedback.","intents":["I want to teach students how to critically evaluate ML systems research papers like conference reviewers do","I need a structured format where every student gets multiple presentation opportunities with different analytical perspectives","I want to create a peer-learning environment where students learn from each other's reviews before group discussion"],"best_for":["graduate-level computer science programs teaching ML systems","instructors building seminar courses with 20-40 enrolled students","research groups training junior researchers in paper evaluation methodology"],"limitations":["Does not scale beyond ~40 students due to fixed presentation slots (each student must present 3+ times)","Requires synchronous participation (Monday 1:00-4:00 PM Pacific time) — no asynchronous alternative documented","No automated review aggregation or conflict-of-interest management (manual instructor coordination)","Relies on manual email/Slack collection of reviews — no structured submission system"],"requires":["UC Berkeley enrollment or guest access to course Slack workspace","Access to assigned research papers (institutional library subscription or manual acquisition)","Synchronous availability for Monday 1:00-4:00 PM Pacific seminars","Zoom client for remote attendance option"],"input_types":["research papers (PDF format)","written reviews (plain text or markdown via email/Slack)","presentation slides (format unspecified)"],"output_types":["written peer reviews (text)","presentation transcripts/notes (text)","instructor feedback (text via Slack/email)","discussion summaries (instructor-generated, format unknown)"],"categories":["planning-reasoning","academic-peer-review"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_1","uri":"capability://memory.knowledge.guest.expert.speaker.integration.for.systems.research","name":"guest-expert-speaker-integration-for-systems-research","description":"Schedules industry and academic experts (e.g., Reynold Xin from Databricks) to present on specific ML systems topics during seminar sessions, with speaker slides and discussion topics posted on course website and Slack. Guest speakers are invited to discuss their research or systems work, providing real-world context for the week's assigned papers. Zoom links are posted to Slack for remote attendance, enabling asynchronous participation for students unable to attend live.","intents":["I want to expose students to cutting-edge ML systems research directly from practitioners and researchers","I need to supplement paper discussions with expert perspectives on implementation trade-offs and real-world constraints","I want to build connections between students and industry/academic leaders in ML systems"],"best_for":["graduate seminars in specialized technical domains (ML systems, distributed computing)","instructors with networks in industry/academia to recruit speakers","programs seeking to bridge academic research and industry practice"],"limitations":["Guest speaker availability is unpredictable and not guaranteed (only Reynold Xin confirmed for week 2 in provided excerpt)","No recorded lectures documented — speakers may not consent to recording or distribution","Synchronous-only delivery (Zoom link posted on Slack, no asynchronous replay mentioned)","Speaker quality and relevance depends entirely on instructor's network and speaker availability"],"requires":["Instructor relationships with industry/academic experts in ML systems","Zoom account for hosting speaker sessions","Slack workspace for distributing Zoom links and speaker materials","Speaker willingness to participate (no compensation model documented)"],"input_types":["speaker presentation slides (format unspecified, likely PDF or Google Slides)","speaker bio/research summary (text)"],"output_types":["live presentation video (Zoom recording, if enabled)","presentation slides (PDF or link)","Q&A discussion notes (instructor-generated, format unknown)"],"categories":["memory-knowledge","professional-development"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_2","uri":"capability://memory.knowledge.course.website.and.reading.list.curation","name":"course-website-and-reading-list-curation","description":"Maintains a static course website (hosted on GitHub Pages at ucbrise.github.io/cs294-ai-sys-sp22/) with weekly reading lists, lecture slides, speaker information, and project guidelines. The website is version-controlled via GitHub, allowing instructors to update readings and materials each semester. Students can suggest additional readings via GitHub pull requests, creating a crowdsourced reading list expansion mechanism.","intents":["I want a centralized, version-controlled repository of course materials that persists across semesters","I need to publish weekly reading assignments and speaker information in a discoverable format","I want to enable students to contribute reading suggestions without direct email coordination"],"best_for":["instructors comfortable with GitHub and static site generation","courses with stable, reusable reading lists across multiple offerings","programs seeking to make course materials publicly discoverable"],"limitations":["Static site generation means no dynamic content (no real-time enrollment updates, no personalized reading recommendations)","GitHub-based workflow requires students to understand pull requests and Git (barrier for non-technical students)","No built-in access control — all course materials are publicly visible (may expose sensitive information if not carefully curated)","Updates require instructor action (no automated scheduling of weekly content releases)"],"requires":["GitHub account and repository access","GitHub Pages enabled on repository","Markdown or HTML knowledge for content updates","Git CLI or GitHub web interface familiarity"],"input_types":["markdown files (course content, reading lists)","YAML frontmatter (metadata, dates, speaker info)","PDF links (lecture slides, papers)"],"output_types":["static HTML website (GitHub Pages)","markdown source files (GitHub repository)","reading list (structured as markdown or YAML)"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_3","uri":"capability://tool.use.integration.slack.based.asynchronous.communication.and.announcements","name":"slack-based-asynchronous-communication-and-announcements","description":"Uses Slack as the primary communication channel for course announcements, guest speaker Zoom links, office hours coordination, and real-time discussion. Instructors post weekly updates, speaker information, and logistical details to Slack channels; students use Slack threads to ask questions and coordinate. Office hours are arranged via email but discussions may occur in Slack channels.","intents":["I want a low-friction way to distribute time-sensitive information (speaker Zoom links, schedule changes) to all students","I need a space for students to ask quick questions without formal email etiquette","I want to create an informal community where students discuss papers and systems outside of class"],"best_for":["graduate seminars with 20-40 students","programs where students already use Slack for other courses or research groups","instructors seeking to reduce email overhead"],"limitations":["Slack messages are ephemeral and not searchable after workspace history limit (free tier: 90 days; paid: unlimited)","Requires Slack account creation and workspace access (barrier for students without Slack experience)","No structured data capture — announcements are unstructured text, making it hard to extract key information programmatically","Slack dependency means course communication is tied to a proprietary platform (not portable if Slack is discontinued or access revoked)"],"requires":["Slack workspace created and managed by instructor or department","Slack account for each student (requires email address)","Slack client (web, desktop, or mobile)","Slack API access for any automation (not documented in course materials)"],"input_types":["text messages (announcements, questions)","Zoom links (pasted as text or links)","file attachments (slides, documents)"],"output_types":["message threads (text)","pinned messages (announcements)","file storage (Slack file uploads)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_4","uri":"capability://code.generation.editing.hands.on.project.delivery.and.evaluation","name":"hands-on-project-delivery-and-evaluation","description":"Requires students to complete a hands-on project (details not fully specified in provided excerpt) that applies ML systems concepts from the course. Projects are evaluated by instructors and may involve implementation, benchmarking, or systems design. The course encourages mixed teams of AI and systems students to collaborate on projects that bridge both domains.","intents":["I want students to apply ML systems concepts to a real problem, not just read papers","I need to assess whether students can design and implement systems-level optimizations for ML workloads","I want to encourage cross-disciplinary collaboration between AI and systems researchers"],"best_for":["graduate programs with access to GPU/TPU clusters or cloud compute credits","courses where students have prior implementation experience (C++, Python, CUDA, etc.)","programs seeking to produce systems-level contributions to ML frameworks"],"limitations":["Project scope and requirements are not documented in provided course materials (unknown complexity, time commitment, evaluation criteria)","Requires access to compute resources (GPUs, TPUs, or cloud credits) — not all students may have equal access","No documented project examples or starter code (students must design projects from scratch)","Evaluation rubric is unknown (no grading criteria provided in excerpt)"],"requires":["Programming experience in Python, C++, or similar systems language","Access to GPU/TPU compute (local, university cluster, or cloud credits)","ML framework familiarity (TensorFlow, PyTorch, etc.)","Systems knowledge (distributed computing, optimization, profiling)"],"input_types":["project proposal (text or slides)","implementation code (Python, C++, CUDA, etc.)","benchmark results (CSV, JSON, or plots)","project report (markdown or PDF)"],"output_types":["project code (GitHub repository)","benchmark results (performance metrics, graphs)","project report (PDF or markdown)","instructor feedback (text via email or Slack)"],"categories":["code-generation-editing","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_5","uri":"capability://tool.use.integration.zoom.based.remote.attendance.option","name":"zoom-based-remote-attendance-option","description":"Provides Zoom links (posted on Slack) for students unable to attend Monday seminars in person, enabling synchronous remote participation in paper discussions and guest speaker sessions. Zoom attendance is optional but encouraged; no asynchronous recording or replay is documented. Remote students can participate in real-time Q&A and discussion.","intents":["I want to accommodate students with scheduling conflicts or physical accessibility needs","I need to enable remote participation without creating a two-tier experience (in-person vs. recorded)","I want to maintain synchronous discussion quality while allowing flexibility"],"best_for":["hybrid-learning programs with geographically distributed students","courses where synchronous discussion is critical (seminars, workshops)","institutions with Zoom licenses for all students"],"limitations":["Synchronous-only (no recorded replay documented) — students in different time zones cannot participate asynchronously","Zoom links are posted on Slack, creating a two-step discovery process (students must check Slack for links)","No documented waiting room, breakout room, or interactive feature usage (standard Zoom, no customization)","Remote students may experience reduced engagement compared to in-person participants (no documented mitigation strategies)"],"requires":["Zoom account with meeting hosting privileges","Zoom client installed on student devices (web, desktop, or mobile)","Stable internet connection (minimum 2.5 Mbps for HD video)","Slack access to receive Zoom links"],"input_types":["Zoom meeting link (URL)","meeting password (if required)"],"output_types":["live video/audio stream (Zoom meeting)","meeting recording (if enabled, format unknown)","chat transcript (Zoom chat, if captured)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-ai-sys-sp22-machine-learning-systems-university-of-california-berkeley__cap_6","uri":"capability://tool.use.integration.email.based.office.hours.coordination","name":"email-based-office-hours-coordination","description":"Instructors offer office hours by email arrangement (no fixed schedule documented), allowing students to request one-on-one meetings to discuss papers, projects, or course content. Students email instructors to schedule meetings; office hours may occur in-person or via Zoom depending on student preference and instructor availability.","intents":["I want to provide individual feedback on student work without scaling office hours to large groups","I need flexibility to accommodate student schedules without committing to fixed office hours","I want to encourage students to seek help on papers and projects they find challenging"],"best_for":["small graduate seminars (20-40 students) where per-student office hours are feasible","instructors with flexible schedules who can accommodate ad-hoc meetings","courses where individual feedback is critical (paper reviews, project guidance)"],"limitations":["No fixed office hours schedule (students must email to arrange, creating coordination overhead)","Scalability issues if many students request office hours simultaneously (no documented queue or scheduling system)","Email-based coordination is asynchronous and slow (may take hours/days to schedule a meeting)","No documented office hours policy (unclear if students are guaranteed access or if there are limits)"],"requires":["Instructor email address (published on course website or Slack)","Student email account","Zoom account (if remote office hours are offered)"],"input_types":["email message (student request for office hours)","proposed meeting times (text in email)"],"output_types":["email confirmation (instructor response with meeting time/location)","Zoom link (if remote meeting)","meeting notes (instructor-generated, format unknown)"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":19,"verified":false,"data_access_risk":"high","permissions":["UC Berkeley enrollment or guest access to course Slack workspace","Access to assigned research papers (institutional library subscription or manual acquisition)","Synchronous availability for Monday 1:00-4:00 PM Pacific seminars","Zoom client for remote attendance option","Instructor relationships with industry/academic experts in ML systems","Zoom account for hosting speaker sessions","Slack workspace for distributing Zoom links and speaker materials","Speaker willingness to participate (no compensation model documented)","GitHub account and repository access","GitHub Pages enabled on repository"],"failure_modes":["Does not scale beyond ~40 students due to fixed presentation slots (each student must present 3+ times)","Requires synchronous participation (Monday 1:00-4:00 PM Pacific time) — no asynchronous alternative documented","No automated review aggregation or conflict-of-interest management (manual instructor coordination)","Relies on manual email/Slack collection of reviews — no structured submission system","Guest speaker availability is unpredictable and not guaranteed (only Reynold Xin confirmed for week 2 in provided excerpt)","No recorded lectures documented — speakers may not consent to recording or distribution","Synchronous-only delivery (Zoom link posted on Slack, no asynchronous replay mentioned)","Speaker quality and relevance depends entirely on instructor's network and speaker availability","Static site generation means no dynamic content (no real-time enrollment updates, no personalized reading recommendations)","GitHub-based workflow requires students to understand pull requests and Git (barrier for non-technical students)","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.14,"ecosystem":0.25,"match_graph":0.25,"freshness":0.5,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.35,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"inactive","updated_at":"2026-06-17T09:51:02.370Z","last_scraped_at":"2026-05-03T14:00:30.220Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","compare_url":"https://unfragile.ai/compare?artifact=ai-sys-sp22-machine-learning-systems-university-of-california-berkeley"}},"signature":"HxySYq/OKkosnuFrEZIZytO5pFQ+hCbdS9RgxnxodTi3JuPC1CHVyR9YsHbDn+7Q99wmOPieO/I5Z9oIE8lWDA==","signedAt":"2026-06-20T08:07:05.426Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","artifact":"https://unfragile.ai/ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","verify":"https://unfragile.ai/api/v1/verify?slug=ai-sys-sp22-machine-learning-systems-university-of-california-berkeley","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}