Research & Analysis AI use cases

Knowledge ManagementResearch & Analysis+1

Internal knowledge assistant workflow where consultants use Lilli to search McKinsey's knowledge base, synthesize key points, surface relevant documents, and find internal experts.

Outcome: The workflow helps consultants move faster from a client question to a grounded starting point, while keeping human judgment in charge of what becomes client advice.

Customer SupportResearch & Analysis+1

AI CEO hotline workflow where customers and merchants talk to an AI version of Klarna's CEO, then transcripts and summaries are analyzed into product feedback for internal teams.

Outcome: The useful result is a faster, more conversational feedback intake loop that routes customer and merchant issues toward product and engineering teams without relying only on surveys or support tickets.

Research & AnalysisKnowledge Management+1

Direct-from-source job indexing workflow that crawls company career pages, summarizes job descriptions with GPT, and enables granular filtering across full job descriptions.

Outcome: Job seekers get a more usable search experience across source-linked listings, while the product team can enrich large volumes of unstructured job descriptions without hand-tagging every role.

Research & AnalysisSales & Outreach+1

Banker client-brief workflow where AI helps prepare briefing documents and surface market/client context so bankers can support more client relationships.

Outcome: The useful result is faster meeting preparation and broader client coverage, while keeping bankers responsible for judgment, advice, and client communication.

Life SystemsPersonal Productivity+1

Uses AI to turn personal routine data and business context into a lightweight reflection loop, while also using synthetic customer profiles to stress-test ideas, messaging, and assumptions.

Outcome: Shows how AI can make scattered personal and business signals easier to think through. The workflow turns logs, notes, habits, and customer assumptions into clearer questions, patterns, and next actions.

Business OperationsResearch & Analysis

Uses AI for high-volume regulated-document work and as a communication coach for product teams, helping organize submission materials, review consistency, and improve stakeholder messaging.

Outcome: Shows how AI can help with document-heavy, detail-sensitive work by turning scattered materials into more structured drafts, checklists, summaries, and communication practice without removing the need for expert human review.

Research & AnalysisCoding & Engineering

Uses error analysis and targeted evals to improve AI products by inspecting failures, grouping them into patterns, and turning those patterns into tests that guide future product changes.

Outcome: Shows a practical way to improve AI quality without guessing. Instead of only tweaking prompts, teams can study real failures, classify them, build focused evals, and use those evals to decide whether the product is actually getting better.

Business OperationsPrototyping & Design+1

Uses a multi-tool AI product management workflow to take a product idea from user research to PRD, clickable prototype, and stakeholder-ready vision video in about 20 minutes.

Outcome: Turns early product discovery into a fast, tangible package: evidence from user discussions, a structured PRD, a working prototype, and a short concept video that stakeholders can react to.

Prototyping & DesignResearch & Analysis

Uses example conversations as the starting point for AI product design, then turns those conversations into clearer requirements, product flows, and interactive prototypes.

Outcome: Shows how teams can design AI products around real conversational behavior instead of jumping straight to UI screens. The workflow helps clarify what users ask for, how the assistant should respond, where trust breaks down, and what the product experience should feel like.

Creative MediaResearch & Analysis

Built AI workflows for documentary post-production at Florentine Films: a searchable media database for archival assets, an iOS field-research app for physical archives, and a Mac OCR tool for difficult historical documents.

Outcome: Turns tedious archival logging into structured, searchable research material. Images, video, audio, physical archive photos, and hard-to-read documents become easier to describe, tag, search, and reuse in documentary storytelling.

Business OperationsResearch & Analysis+1

Uses AI across recruiting, product research, and personal logistics: finding strong candidates on LinkedIn, creating interactive customer personas, and building a hyperlocal helper for parking decisions.

Outcome: Shows how AI can support three different kinds of judgment-heavy work: identifying people worth reaching out to, making customer research easier to interrogate, and turning local context into a practical decision aid.

Knowledge ManagementResearch & Analysis+1

Built a personalized Claude Code system for task management, automated academic research, and modular context files that make AI collaboration easier with shorter, simpler prompts.

Outcome: Turns Claude Code into a personal operating layer for knowledge work: daily tasks are generated from Markdown files, research appears as a digest, papers are summarized automatically, and context files help Claude give better feedback without long prompts.

Research & AnalysisMarketing & Content+1

Uses an AI-assisted research workflow to digest dozens of weekly podcast episodes into structured notes, patterns, insights, and blog-ready ideas for investing and writing.

Outcome: Shows how AI can turn a high-volume information habit into a repeatable research system. Instead of relying on memory or scattered highlights, Tomasz can process many episodes, extract themes, and turn signal into writing or investment thinking.