161 lines
38 KiB
JSON
161 lines
38 KiB
JSON
{"prompt": "ok so ive been thinking about our ability system in unreal and i keep coming back to gameplay ability system vs rolling our own. GAS is a lot of machinery and the team doesnt know it, but our own thing already has replication bugs we dont understand. write up the honest comparison for a 4 person team shipping in a year, and pick", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "server rpc validation for the interact action — range check, line of sight, and cooldown, all server side, and log rejections with the actor", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "radial menu for emotes, controller stick to select, snap to 8 sectors, and it should show which one is selected clearly at a glance", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the interact range is 200 units, designers want 350", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "our actor components each hold a raw pointer back to the owner and cast it. clean that up with proper interfaces, same behavior in the test map", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "clients desync from the server after about 20 minutes and the only symptom is doors being open for some players and closed for others", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "whats our current replication cost per actor, roughly, based on the properties marked replicated? just read it and tell me", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the design doc for the loot system from what's already implemented, because nobody can tell whats intentional anymore", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "rails: background job to recalculate the leaderboard every 5 minutes, sidekiq, and it must be safe if two enqueue at once", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "our models have 40 scopes and half of them are unused. find the dead ones and remove them, nothing else changes", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "n+1 queries all over the profile page. bullet is screaming. sort it out", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "plan the upgrade from rails 6.1 to 7.2. 900 files, a lot of custom middleware, and we're on webpacker", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain what our ApplicationRecord callbacks do on save, in order, because something is touching updated_at that shouldnt", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "requests time out on one endpoint after we added an index, which makes no sense to me", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "api documentation for the public endpoints, from the controllers and serializers as they actually are, not what the old doc claims", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "svelte: the settings page needs optimistic toggles that revert on failure with a small inline error, no toasts", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "astro build outputs a 900kb bundle for a mostly static site, and i cant tell what's pulling in what", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "focus outline is `outline: none` in the global reset", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "we have three ways of doing modals in the app — a store-driven one, a context one, and one that just uses `<dialog>`. pick the dialog one and port the rest, same behavior including focus handling", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "sketch out the plan for our marketing site rebuild — content in mdx, previews per branch, and the design team editing copy without a PR", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "content collection schema for case studies, with the frontmatter validated at build time so a missing field fails the build", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "is the view transition api actually doing anything on our routes or are we just paying for the code? tell me what you see", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "docs page listing our css custom properties and what they control, for people theming the widget", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "hero animation on the landing page: staggered fade-up on load, and it must not run again on client-side navigation back to home", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "go", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "so we've got this thing where our rails app and the new go service both write to the users table and i hate it. i want a plan to get the go service to stop writing directly — an api, or events, or just moving the ownership entirely. cost out each one, we have maybe 6 weeks of appetite for this", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "internal api on the rails side for user reads and writes so the go service can stop touching the table, with a token per consumer", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "admin ui to see which service last wrote each user record, from the audit trail", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "the go service's db user has full write grants, restrict it to read for now", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "both services define the user shape independently. generate one from the other so they cant drift, no field changes today", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "a user's email gets reverted to the old value occasionally, minutes after they change it. two writers obviously but i want to know exactly which path does it", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "compare how the two services validate an email address and tell me which is stricter", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "adr documenting who owns the user record after this change, and the rule for future services", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "here's the log from the failing job, i dont even know where to start:\n\nI, [2026-07-29T02:14:08.221Z] INFO -- : [Sidekiq] start ReindexJob jid=8a2f11 args=[\"batch\", 4412]\nW, [2026-07-29T02:14:39.902Z] WARN -- : [Sidekiq] ReindexJob jid=8a2f11 PG::TRDeadlockDetected\nI, [2026-07-29T02:14:39.905Z] INFO -- : [Sidekiq] fail ReindexJob jid=8a2f11 elapsed=31.68\nI, [2026-07-29T02:14:40.001Z] INFO -- : [Sidekiq] start ReindexJob jid=8a2f11 args=[\"batch\", 4412] retry=1\nW, [2026-07-29T02:15:11.442Z] WARN -- : [Sidekiq] ReindexJob jid=8a2f11 PG::TRDeadlockDetected\n...\nI, [2026-07-29T02:41:03.118Z] INFO -- : [Sidekiq] fail ReindexJob jid=8a2f11 retry=24 dead=true\n\nit deadlocks with itself as far as i can tell, theres only one worker on that queue", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "batch reindex job that processes in stable id order with a bounded transaction per chunk, so it cant deadlock against itself", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the sidekiq concurrency is 25 against a pool of 10 connections", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "job queue page: per-queue depth, latency, and the dead set with a retry-selected action", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "our jobs take positional args in some cases and hashes in others, and a couple take activerecord objects which is asking for trouble. standardize on ids, no change in what each job does", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan how we'd move sidekiq jobs to a separate service so a bad job cant take down the web pods", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "read our job retry configuration and tell me which jobs will retry forever", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the job authoring guide: idempotency, argument rules, retry expectations, and how to make a job safe to run twice", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "unreal: the loading screen needs to not be a black frame for 2 seconds before the level streams in", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the pak file mount order means our patch content loses to the base content", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan the content patching pipeline and then build the manifest diffing tool", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.8, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "work out whats causing the hitching on level load and then write it up in the perf doc", "purpose": "debugging", "secondary": "writing", "mixed": true, "difficulty": 0.8, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "nah do it the other way", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "im going round in circles on this. we need to support 'projects' having many 'environments' and each environment having its own config, secrets and deploy history. right now project and environment are the same table with a nullable parent id and its a nightmare. i want the data model and the migration path written out, including what happens to existing rows and the api compat story", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "environments crud with config inheritance from the project, and an override indicator per key", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "environment switcher in the header, keyboard accessible, with the current env's color as a subtle accent so you know where you are", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "prod environment badge is green, should be red", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the config resolution logic exists in the api, the cli and the deploy worker separately. one implementation, and the resolved config must be identical for every existing environment", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "config changes apply to the wrong environment about once a week according to the audit log. i cannot reproduce it and its terrifying", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "tell me how the config inheritance resolves when a key is set at both levels and one is an empty string", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "docs for the config api — precedence rules, the reserved key prefixes, and what a null vs empty value means", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "deploy history endpoint with the diff of config between deploys, paginated", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "deploy timeline ui: rows per deploy, status, duration, who, and an expander showing the config diff", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the relative timestamps say '2 hours ago' for things that happened 2 minutes ago in another timezone", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "figure out the plan for rollbacks. currently 'deploy the old sha' which doesnt cover config or migrations", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "review the deploy worker for what happens if it dies halfway through, and whether we can end up with a half-applied config", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "deploys report success but the pods still run the old image about 5% of the time", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the rollback runbook including the config and migration parts, and what we tell customers during one", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "consolidate the deploy status enum — the api says 'in_progress', the worker says 'running' and the ui maps between them. pick one, keep the api's public strings stable", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "unser Deploy-Webhook feuert zweimal pro Deploy. plan zuerst wie wir das sauber lösen, dann bau es", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.6, "slice": "mixed", "lang": "de"}
|
|
{"prompt": "have a look at the config resolver and tidy the naming as you go", "purpose": "review", "secondary": "refactor", "mixed": true, "difficulty": 0.5, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "do the needful", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "we're adding a public terraform provider. i want the design first — resource coverage for v1, import support, how we handle our api's eventual consistency in the read-after-create, and the release process for the registry", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "terraform provider resource for projects and environments, with import, and retries on the read after create until it's consistent", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "provider docs — one page per resource, every attribute, an import example, and the note about eventual consistency", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the provider version constraint in our examples pins an old version", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "our api client in the provider duplicates the one in the cli. share it as a library, both keep working identically", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "terraform plan shows a change on every run for one attribute even when nothing changed", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "does our provider handle a resource deleted outside terraform gracefully, or does it error on refresh", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "a small web ui for browsing our resource schemas, generated from the provider's schema json", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan the sdk story — do we hand write three sdks or generate from openapi. include the maintenance cost and how idiomatic the generated ones would be", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "typescript sdk with typed responses, automatic pagination via async iterators, and a pluggable fetch", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the sdk readme example doesnt compile, wrong import path", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.1, "slice": "core", "lang": "en"}
|
|
{"prompt": "sdk reference docs — every method, params, the error types, and a section on retries and timeouts", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the python and ts sdks name the same concepts differently (client vs session, list vs iterate). align the naming across both, keeping the old names as aliases", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "the sdk hangs on a 429 instead of backing off, but only when the retry-after header is a date not a number", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "read the sdk's pagination and tell me what happens if a page boundary lands on a deleted record", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "docs site nav is 40 items flat, group it and add a search", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "later gator", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "ok real talk our monitoring for the game servers is basically nothing. we know when a match server crashes because players complain. i want the plan: what we instrument, how we get it off the boxes, and the three alerts that would actually have caught the incidents we've had", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "match server heartbeat and metrics reporting — tick time, player count, memory, sent every 10s to the fleet manager", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "fleet view: servers as tiles by region, color by health, click for the match currently running on it", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the heartbeat timeout is 5s and gc pauses are 6s", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "our server allocation code has three copies for the three regions with the hostnames swapped. one implementation, region as a parameter", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "match servers occasionally report 0 players while a match is clearly running, and then recover", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain how a crashed match server is detected and what happens to the players in it right now", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "ops doc for the game server fleet — how to drain a server, how to roll a build, and what the alerts mean", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "matchmaking backfill so a match that loses players can get topped up from the queue, with a skill window that respects the match in progress", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "queue ui showing estimated wait and your current search range widening, without lying to the player", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "queue estimate shows 'less than a minute' always", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "map out how we'd add a ranked mode. placement matches, decay, seasons, and the leaderboard consistency story", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "one region matches players 400 mmr apart and the others dont. same config as far as i can see", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "read the mmr update code and tell me if a player can lose rating for a win in any edge case", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "player-facing explainer for how ranked works, honest about what mmr is and isn't", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "the mmr math lives in the match server and the leaderboard service, separately implemented. one implementation, same ratings on the historical match set", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "现在的反作弊只有客户端检测。先给我方案，服务端能验证哪些行为，误报怎么处理", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "zh"}
|
|
{"prompt": "server side speed and teleport detection with a tolerance for latency, flagging not banning, and a review queue", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "review queue ui for flagged players: the evidence, a replay link, and actions with a required note", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the flag threshold is 1 event which flags everybody with bad wifi", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "core", "lang": "en"}
|
|
{"prompt": "legit players get flagged when their connection hiccups, and i cant tell from the code whether we account for the resync at all", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "explain the current client-side checks and what a determined cheater would have to do to bypass each one", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "internal policy doc for enforcement actions — what triggers a warning, a suspension, a ban, and the appeal path", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "the anti-cheat report parsing duplicates the telemetry parser with a different struct. share the parsing, same reports produced", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "yeah that", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "need to decide how we store timeseries for the device fleet. 30k devices, one reading a minute, 18 month retention, and the queries are 'last value per device' and 'hourly average over a month'. timescale, clickhouse, or roll our own on postgres partitions. write it up with numbers", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "continuous aggregate for hourly averages with a refresh policy, and the last-value query needs to be fast without scanning", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "device detail page with a sparkline per metric, a range picker, and a live-updating current value", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the retention policy drops chunks at 12 months not 18", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "our downsampling job and the continuous aggregate compute averages differently for partial hours. pick the aggregate's behavior and delete the job", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "hourly averages are wrong for the hour containing a dst transition, in one direction only", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "check whether our ingest path can drop readings silently under load — i see a bounded channel with a default case", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "write up the data model and query patterns for the fleet timeseries so people stop writing full scans", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "alerting on device metrics: thresholds per device type, a sustained-for duration, and hysteresis so it doesnt flap", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "alert config form: metric picker, comparator, threshold, duration, and a preview of how often it would have fired last week", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the alert email says 'threshold exceeded' with no metric name or value", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan the firmware telemetry schema so we can add fields without breaking old devices or the ingest path", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "a device that goes offline stops triggering its 'no data' alert after 24 hours, which is backwards", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain the hysteresis in our alert evaluator, i want to know the exact conditions for clearing", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "customer facing docs for device alerts, what each metric means physically, and the sampling caveat", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "our alert state machine is spread across the evaluator, the notifier and a boolean column. make the state explicit in one place, same firing behavior on the replayed history", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "sure", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "pasted from the ticket, this is what the customer sent us:\n\n\"When we PATCH /v2/devices/{id} with only the `label` field, the response comes back with\n`tags: []` and our tags are gone. Sending the same request with `tags` included works.\nWe were told PATCH was a partial update. We've lost tags on about 900 devices.\nAlso: the response 200s so we had no idea until a customer of ours noticed.\"\n\nthis is bad. i need to know whether its a bug in the handler or the serializer, and what we do about the 900 devices", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "pasted-context", "lang": "en"}
|
|
{"prompt": "proper patch semantics on the device endpoints — absent means untouched, explicit null means clear, and reject unknown fields", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the api docs describe PATCH as a partial update, which the code doesnt do. fix the docs for now and mark the discrepancy clearly", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "device edit form should only send changed fields", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "our request models use zero values as 'not provided' which is exactly this bug in seven other endpoints. move to explicit optionals everywhere, behavior fixed consistently", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "audit the other write endpoints for the same absent-vs-empty confusion and list them", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "restore script for the 900 devices' tags from the audit log, dry run first", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "write the customer apology and the status page note, and be specific about what we're doing", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "plan how we prevent this class of bug — contract tests, or codegen from the spec, or a review checklist. pick something enforceable", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "contract tests generated from the openapi spec, run in ci, failing on any response that doesnt match", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "the spec's example for the device object is missing three fields the api returns", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "look at our error response shapes across endpoints and tell me how many distinct shapes we actually emit", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "unify the error response shape across all endpoints, keeping the status codes and messages identical to today", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "validation errors return 400 on one endpoint and 422 on another for the same kind of problem", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "error reference page for the api: code, http status, meaning, and whether retrying will help", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "toast for api errors that shows the request id so support can find it, without being ugly about it", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "core", "lang": "en"}
|
|
{"prompt": "keep at it", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.3, "slice": "vague-eval", "lang": "en"}
|
|
{"prompt": "design the plugin marketplace: submission, review, versioning, and the permission model for what a plugin can access. i want the security boundary spelled out precisely because this is the part we cant get wrong", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "plugin permission prompts at install time, enforced at runtime, and a way for the user to see what a plugin has accessed", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "marketplace browse page: cards with icon, name, author, install count, and a filter by permission required", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "install count shows 0 for everything, its reading the wrong field", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "the plugin host and the extension host are two implementations of the same sandbox with different capabilities. merge them, and enumerate any capability that changes for existing plugins", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.9, "slice": "core", "lang": "en"}
|
|
{"prompt": "one plugin can read another plugin's stored data and i dont know if thats by design or a bug", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "assess our sandbox: what can a malicious plugin actually do today, concretely", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "plugin developer docs — the api surface, the permission declarations, the review criteria, and what gets you rejected", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "boundary", "lang": "en"}
|
|
{"prompt": "plugin update mechanism with a changelog shown to the user, and permission changes requiring re-consent", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.7, "slice": "core", "lang": "en"}
|
|
{"prompt": "installed plugins list with enable/disable toggles, an update badge, and a settings link per plugin", "purpose": "frontendImpl", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "disabled plugins still run their background handlers", "purpose": "debugging", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "the version comparison uses string sort so 1.10.0 < 1.9.0", "purpose": "quickFix", "secondary": null, "mixed": false, "difficulty": 0.2, "slice": "core", "lang": "en"}
|
|
{"prompt": "map out the revenue share and payouts side of the marketplace, including tax and refunds, at a design level", "purpose": "planning", "secondary": null, "mixed": false, "difficulty": 0.8, "slice": "core", "lang": "en"}
|
|
{"prompt": "explain what happens today if a plugin author deletes their published version that people have installed", "purpose": "review", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "marketplace listing guidelines and the review rubric, written so a reviewer and an author read it the same way", "purpose": "writing", "secondary": null, "mixed": false, "difficulty": 0.5, "slice": "core", "lang": "en"}
|
|
{"prompt": "our manifest parsing accepts three schema versions with a chain of if-statements. make it a proper versioned parser, same manifests accepted", "purpose": "refactor", "secondary": null, "mixed": false, "difficulty": 0.6, "slice": "core", "lang": "en"}
|
|
{"prompt": "figure out the sandbox design and then implement the permission check layer", "purpose": "planning", "secondary": "backendImpl", "mixed": true, "difficulty": 0.9, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "explain how plugin storage isolation works then write the section for the developer docs", "purpose": "review", "secondary": "writing", "mixed": true, "difficulty": 0.7, "slice": "mixed", "lang": "en"}
|
|
{"prompt": "and the rest", "purpose": "backendImpl", "secondary": null, "mixed": false, "difficulty": 0.4, "slice": "vague-eval", "lang": "en"}
|