{ "version": "0.1.0", "generatedAt": "2026-05-16T00:00:00+05:30", "purpose": "Candidate-ready application packet and work-sample schema for Edxperimental Labs hiring.", "contacts": [ { "name": "Sanjay Prasad", "email": "sanjay@edxperimentallabs.com", "role": "Research direction, benchmark design, and technical delivery" }, { "name": "Saujas", "email": "saujas@edxperimentallabs.com", "role": "Sales engineering, client discovery, and solution design" } ], "roleCatalog": [ { "role": "AI research analyst", "status": "Open to exceptional candidates", "owner": "Sanjay Prasad", "mission": "Turn primary sources, model releases, benchmark papers, and pricing changes into buyer-readable research.", "firstArtifacts": [ "Source-backed research note", "Chart-ready source table", "Model/provider decision memo" ], "proofSignals": [ "Primary-source discipline", "Clear uncertainty labels", "One useful visual", "Buyer decision" ], "trialProject": "Publish a short research note on an AI benchmark, model release, or cost curve with one original chart and a weakest-claim note.", "readiness": 82 }, { "role": "Benchmark engineer", "status": "Project-based first", "owner": "Sanjay Prasad", "mission": "Convert messy workflows into task packets, trace schemas, scoring rubrics, replay artifacts, and inspection pages.", "firstArtifacts": [ "Task packet", "Scoring rubric", "Trace schema", "Replay checklist" ], "proofSignals": [ "Task realism", "Leakage control", "Reviewer-friendly scoring", "Runnable verification" ], "trialProject": "Design one benchmark task and artifact bundle for an agent, browser, support, security, or Indian-workflow suite.", "readiness": 78 }, { "role": "Sales engineer", "status": "Selective outreach", "owner": "Saujas", "mission": "Translate buyer problems into the smallest benchmark sprint, demo path, and evidence request that can answer the decision.", "firstArtifacts": [ "Discovery memo", "Buyer question map", "Sprint scope", "Demo handoff" ], "proofSignals": [ "Client diagnosis", "Scope discipline", "Technical translation", "Next-action quality" ], "trialProject": "Write a discovery memo for a hypothetical buyer and convert it into a one-week benchmark sprint with owner routing.", "readiness": 76 }, { "role": "Studio product builder", "status": "Project-based first", "owner": "Sanjay Prasad", "mission": "Turn repeated research and consulting workflows into usable Studio demos, calculators, dashboards, and product packets.", "firstArtifacts": [ "Interactive demo", "Product packet", "Verification screenshot", "Demo readiness notes" ], "proofSignals": [ "Useful interface", "Responsive polish", "Evidence-linked controls", "Clear handoff" ], "trialProject": "Build a small Studio interface that makes one benchmark, cost curve, or service decision easier to inspect.", "readiness": 72 } ], "roleTracks": [ { "title": "AI research analyst", "status": "Open to exceptional candidates", "prompt": "Write a source-backed memo on one model, benchmark, provider, or cost curve. Include one original chart, three primary sources, and a buyer decision.", "evaluation": [ "Source quality", "Chart clarity", "Decision usefulness", "Failure-mode honesty" ] }, { "title": "Benchmark engineer", "status": "Project-based first", "prompt": "Design a workflow benchmark packet with expected output, hidden checks, scoring rubric, trace schema, and replay plan.", "evaluation": [ "Task realism", "Rubric precision", "Replayability", "Leakage control" ] }, { "title": "Sales engineer", "status": "Selective outreach", "prompt": "Prepare a discovery memo for a client evaluating AI in one workflow. Convert the problem into a one-week benchmark sprint and handoff plan.", "evaluation": [ "Client diagnosis", "Scope discipline", "Technical translation", "Next-step quality" ] } ], "rubric": [ { "dimension": "Evidence taste", "strongSignal": "Cites primary sources, shows uncertainty, and keeps claims attached to inspectable proof." }, { "dimension": "Builder speed", "strongSignal": "Turns an idea into a memo, harness, analysis page, script, or demo without waiting for perfect instructions." }, { "dimension": "Buyer clarity", "strongSignal": "Explains the decision, the risk, the tradeoff, and the next test in language an operator can use." }, { "dimension": "Scope discipline", "strongSignal": "Chooses a narrow useful artifact instead of pretending to solve a whole category in one pass." }, { "dimension": "Aesthetic judgment", "strongSignal": "Ships work that feels polished enough for a serious AI buyer to trust and share internally." } ], "requiredFields": [ "candidateName", "email", "roleTrack", "workSampleType", "artifactUrl", "summary", "sourceLinks", "availability" ], "optionalFields": [ "github", "linkedin", "portfolio", "location", "notes" ], "artifactTypes": [ { "name": "Benchmark note", "description": "A short critique of SWE-bench, BrowserGym, HELM, LiveBench, OSWorld, tau-bench, or an enterprise workflow benchmark." }, { "name": "Cost model", "description": "A spreadsheet, notebook, or code snippet estimating cost per accepted output for a real AI workflow." }, { "name": "Workflow task", "description": "A task packet with expected output, evidence requirement, scoring rubric, and likely failure modes." }, { "name": "Studio demo", "description": "A small interface, notebook, or script that makes an AI evaluation easier to inspect." } ], "hiringPipeline": [ { "stage": "Artifact screen", "owner": "Sanjay Prasad", "evidence": "One work sample, source trail, summary, and role track.", "decision": "Does the artifact show evidence taste, builder speed, and buyer clarity?" }, { "stage": "Technical conversation", "owner": "Sanjay Prasad", "evidence": "Walkthrough of assumptions, sources, failure modes, and what the candidate would improve.", "decision": "Can the candidate reason from evidence instead of reciting AI discourse?" }, { "stage": "Client-simulation screen", "owner": "Saujas", "evidence": "A five-minute explanation of the work to a buyer plus a proposed next sprint.", "decision": "Can the candidate translate technical details into a scoped business decision?" }, { "stage": "Trial artifact", "owner": "Sanjay Prasad and Saujas", "evidence": "A paid or project-based artifact shipped against a narrow brief.", "decision": "Does the work fit the publishing, Studio, benchmark, or consulting operating loop?" } ], "first90DayPlan": [ { "window": "Days 1-15", "focus": "Orientation through shipped artifacts", "outputs": [ "One source-backed research note", "One benchmark critique", "One internal operating note" ] }, { "window": "Days 16-45", "focus": "Own one public surface", "outputs": [ "One article or Studio packet", "One chart-ready source table", "One review-ready benchmark/task packet" ] }, { "window": "Days 46-75", "focus": "Turn a client question into a reusable asset", "outputs": [ "Discovery memo", "Sprint scope", "Reusable template, report, or demo surface" ] }, { "window": "Days 76-90", "focus": "Publish and hand off", "outputs": [ "Public artifact", "Maintenance checklist", "Next-measurement backlog" ] } ], "reviewerQuestions": [ "What primary source would change your conclusion?", "Which claim in your artifact is weakest, and what proof would fix it?", "How would this become a Studio product, benchmark task, article, or consulting deliverable?", "What would you remove if a buyer had only five minutes?", "What should Sanjay review and what should Saujas review?" ] }