Enterprise AI Asset Map
17 enterprises mapped across 9 AI asset classes. From atomic skills to autonomous agents — see who needs what.
17
Enterprises
144
Confirmed
9
Potential
100%
Coverage
Stacks:ConfirmedPotential
| Build-time | Deploy-time | Runtime | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Enterprise | Industry | Source | Skills Atomic | Plugins Bundles | MCP Servers Connectors | Workflows Recipes | Eval Suites Quality Gates | Context Packs Domain Knowledge | Intelligence Feeds Live Data | Autonomous Agents Buyers | JACP Protocol Transaction Layer |
| Novo Nordisk | Pharma | Enterprise Guide | Clinical document generation; report formatting; m… | Clinical Study Report bundle: data extraction + me… | Clinical trial management systems, regulatory subm… | CSR pipeline: trial data ingestion → narrative gen… | Output quality vs. manual CSR benchmarks; regulato… | Drug trial protocols, ICH guidelines, therapeutic … | Active trial data streams, adverse event databases… | Autonomous CSR drafting agents that pull trial dat… | High-stakes domain: eval-before-deploy critical; r… |
| Pfizer | Pharma | Enterprise Guide | Document search/summarization (300K docs in 6 week… | Legal document processing suite: search + summariz… | AWS Bedrock integration; internal document reposit… | Automated legal review pipeline: ingest 300K docum… | Summarization accuracy vs. manual review; legal co… | Pharmaceutical regulatory corpus (FDA, EMA); inter… | Regulatory filing deadlines; patent expiration cal… | Autonomous document triage agents that process inc… | Cost reduction focus (infra costs down 55%); quali… |
| LexisNexis | Legal Tech | Enterprise Guide | Automated legal document summarization; classifica… | Legal processing bundle: summarization + classific… | Claude in Amazon Bedrock; internal legal database … | 300K documents/year ingested → AI summarized → cla… | Summary accuracy benchmarks; citation verification… | Case law databases; statutory frameworks per juris… | New case filings; legislative updates; regulatory … | Autonomous legal research agents that monitor new … | Model safety cited as primary selection driver |
| DoorDash | Logistics | Enterprise Guide | Voice-operated self-service contact center; Dasher… | Contact center suite: voice interface + knowledge … | Delivery management system, Dasher app data, order… | Voice self-service pipeline: Dasher query → intent… | First contact resolution rate; average time to res… | Dasher onboarding materials; delivery protocol doc… | Real-time order status; surge pricing data; market… | Voice-operated self-service agent already deployed… | Rapid deployment (2 months); quality monitoring cr… |
| WPP | Advertising | Enterprise Guide | Creative content generation; process enhancement; … | Creative production bundle: ideation + copywriting… | AWS integration; internal creative asset managemen… | Campaign creation pipeline: brief intake → creativ… | Brand guideline adherence; creative quality scorin… | Client brand guidelines; industry creative benchma… | Social media trend signals; campaign performance m… | Autonomous campaign optimization agents that monit… | Innovation-driven adoption ('push boundaries of cr… |
| Lonely Planet | Travel / Media | Enterprise Guide | Travel experience/itinerary generation; destinatio… | Content production bundle: destination research + … | Destination databases; booking system APIs; local … | Content pipeline: destination research → experienc… | Content accuracy (factual correctness of destinati… | Destination guides (historical + current); local c… | Live travel advisories; seasonal event calendars; … | Writers freed for discovery ('finding the next in-… | Cost-driven (80% reduction); freshness of context … |
| Coinbase | Crypto / Finance | Enterprise Guide | Employee productivity tools; customer support augm… | Support + productivity bundles segmented by role | Trading platform APIs; compliance systems; custome… | AI support for every customer + every employee — f… | Hours freed per employee per week; customer resolu… | Cryptocurrency market knowledge; regulatory framew… | Real-time crypto prices; blockchain transaction da… | Vision: AI agent for every customer + every employ… | Financial services = high-risk domain; eval-before… |
| iFood | Food Delivery | Enterprise Guide | Fine-tuned model capabilities on Bedrock; cost-opt… | Order processing + customer support + restaurant m… | Amazon Bedrock integration; restaurant partner sys… | Fine-tuning workflow: base model → domain training… | Cost per inference vs. traditional models; accurac… | Restaurant menus/catalogs; delivery zone maps; cui… | Real-time restaurant availability; driver location… | Autonomous order routing agents; dynamic pricing o… | Cost-efficiency is primary lens ('if you use it co… |
| GitLab | DevTools | Enterprise Guide | Code generation; technical discussion facilitation… | DevOps suite: code generation + review + testing +… | GitLab platform APIs; CI/CD pipeline connectors; c… | Multi-step agentic development: task intake → code… | 5-10% improvement in accuracy, comprehensiveness, … | Repository codebases; GitLab product documentation… | CI/CD pipeline status; merge request activity; sec… | Multi-step agentic software development already in… | Measurable benchmarks (5-10% improvements); regres… |
| Thomson Reuters | Legal / Finance | Enterprise Guide | Industry-expertise-optimized AI; fine-tuned domain… | Legal + financial information processing bundles t… | Internal legal/financial databases; news wire serv… | Domain-optimized pipeline: query intake → fine-tun… | Quality benchmarks against human expert output; sp… | Legal case law corpus; financial reporting standar… | Breaking legal/financial news; regulatory filing a… | Autonomous legal research agents; financial analys… | Fine-tuning as quality mechanism ('tailor AI assis… |
| FeatherSnap | Consumer IoT | Enterprise Guide | Bird species identification; image classification;… | Smart feeder bundle: image recognition + species c… | Amazon Bedrock integration; device firmware APIs; … | IoT inference pipeline: camera capture → image upl… | Species identification accuracy; latency from capt… | Bird species databases; regional migration pattern… | Seasonal migration data; weather conditions affect… | Embedded agent on IoT device: capture → classify →… | Rapid deployment (90 days); low-risk domain but hi… |
| Cognition (Devin) | AI DevTools | Eng Blog: Evals | Autonomous software engineering; full-stack code g… | Full-stack development suite: code gen + testing +… | Code repository access; CI/CD systems; cloud deplo… | Autonomous coding workflow: task spec → planning →… | Referenced in Anthropic eval guide; SWE-bench perf… | Repository codebases; language/framework documenta… | CI/CD pipeline results; test suite outputs; deploy… | Devin IS an autonomous agent — sits at top of stac… | Autonomous agent needs programmatic skill manageme… |
| Bolt AI | AI DevTools | Eng Blog: Evals | Web application generation; full-stack scaffolding… | Web app generation bundle: frontend + backend + da… | Cloud hosting APIs; database provisioners; package… | App generation pipeline: user prompt → architectur… | Referenced as Anthropic eval customer; app functio… | Web framework documentation; UI component librarie… | Framework update notifications; security advisory … | The product itself is an agent that generates and … | Generated apps need quality gates; try-before-buy … |
| Sierra | Customer Experience | Eng Blog: Evals | Customer service conversation handling; intent cla… | Customer experience bundle: conversation + knowled… | Client CRM systems; order management; payment proc… | CX automation pipeline: customer query → intent cl… | Referenced as Anthropic eval customer; CSAT scores… | Client-specific product catalogs; return/refund po… | Real-time order status; inventory levels; promotio… | CX agents operating autonomously with customer-fac… | High-risk customer-facing domain; eval-before-depl… |
| Stripe | Fintech | Eng Blog: Evals | Code generation; API integration; developer docume… | Developer productivity bundle: code gen + API test… | Stripe API ecosystem; internal code repositories; … | Developer workflow: feature spec → code generation… | Referenced as Anthropic eval customer; API correct… | Stripe API reference; payment industry regulations… | API deprecation notices; regulatory changes; fraud… | Autonomous integration agents that help developers… | Financial domain = high compliance requirement; ev… |
| Shopify | E-commerce | Eng Blog: Evals | Code generation; merchant tool development; storef… | Merchant tools bundle: storefront gen + analytics … | Shopify platform APIs; merchant data stores; payme… | Merchant support pipeline: merchant request → capa… | Referenced as Anthropic eval customer; merchant sa… | Shopify platform documentation; Liquid templating … | Platform update notifications; merchant performanc… | Autonomous store optimization agents; inventory ma… | E-commerce domain growing in agent autonomy; merch… |
| Descript | Media / Creative | Eng Blog: Evals | Audio/video transcription; content editing; format… | Media editing bundle: transcription + editing + ef… | Media processing APIs; cloud storage; publishing p… | Content production pipeline: media import → transc… | Referenced as Anthropic eval customer; transcripti… | Media format specifications; audio processing tech… | Platform feature updates; media format standard ch… | Autonomous content post-production agents that han… | Creative output quality is subjective — eval suite… |
Pharma / Life Sciences
Novo Nordisk, Pfizer, Thomson Reuters
Richest fill across all 9 columns. High-stakes, regulation-heavy domains need every layer. Premium customers.
DevTools / AI-native
Cognition, Bolt, GitLab, Stripe, Shopify
Their products ARE agents. They sit in Autonomous Agents AND consume all lower layers. Both buyers and resellers.
Customer-facing Ops
DoorDash, Sierra, Coinbase
Strongest monitor-step requirements. Agents take irreversible actions (refunds, voice responses) — maps to 0.8% irreversible action rate.
Media / Content
Lonely Planet, WPP, Descript
Cost-optimization buyers (80% reduction headline). Eval suites harder because quality is subjective — need model-based judging.