Perplexity AI

@perplexity what are the latest FDA approvals for GLP-1 drugs in the past 6 months, and which manufacturers are involved?

Based on recent regulatory filings and news sources, the FDA approved tirzepatide (Zepbound, Eli Lilly) for chronic weight management in November 2023, and granted expanded indications for semaglutide (Wegovy, Novo Nordisk) in March 2024 to reduce cardiovascular risk in adults with obesity. A third approval came in April 2024 for oral semaglutide (Rybelsus) at higher doses. All three drugs target the GLP-1 receptor pathway, with tirzepatide also activating GIP receptors. This example illustrates how Perplexity synthesizes information from multiple authoritative sources to answer time-sensitive research questions.

This example highlights Perplexity's strength in real-time web search and synthesis across regulatory databases, news, and medical sources. The MCP returns text summaries; it does not provide citations in a structured format. API rate limits apply based on your Perplexity plan tier, and responses reflect the model's training cutoff plus live search results.

@perplexity compare the pricing and feature differences between Stripe, Paddle, and Lemon Squeezy for a SaaS product selling to EU customers

Stripe charges 2.9% + €0.25 per transaction in the EU and requires you to handle VAT compliance separately (though Stripe Tax is available as an add-on). Paddle acts as merchant of record, handling VAT automatically, and charges 5% + €0.50 per transaction, which includes tax remittance. Lemon Squeezy also serves as merchant of record with similar VAT handling, pricing at 5% + €0.50, and offers built-in affiliate management. For EU sales, Paddle and Lemon Squeezy reduce compliance overhead but cost more per transaction. This example demonstrates how Perplexity compares commercial products using current pricing data from vendor sites.

This showcases Perplexity's ability to perform comparative research across vendor documentation and pricing pages. The MCP pulls live web data, so pricing accuracy depends on when vendors last updated their sites. Responses are text summaries, not structured comparison tables. Useful for quick competitive analysis before diving into vendor docs directly.

@perplexity summarize the key arguments from the top 5 Hacker News discussions about Rust vs Go for backend services in 2024

The dominant themes in 2024 HN threads are: (1) Rust's memory safety guarantees justify the steeper learning curve for systems-level services where performance and correctness are critical; (2) Go's simplicity and faster compile times make it preferable for CRUD APIs and microservices where developer velocity matters more than raw speed; (3) both ecosystems have matured—Tokio and Axum for Rust async, Go 1.22 generics closing the expressiveness gap; (4) hiring remains easier for Go; (5) Rust's ownership model prevents entire bug classes but increases cognitive load during refactoring. This example shows Perplexity aggregating community sentiment from a specific platform over a defined time window.

This example demonstrates Perplexity's ability to synthesize discussion threads and extract thematic patterns from community forums. The MCP does not return links to individual HN threads—it summarizes sentiment across search results. Accuracy depends on Perplexity's search index freshness and the model's interpretation of nuanced technical debates. Best used for directional insight, not authoritative citation.

When Perplexity beats your internal wiki for support agents

A 6-person support team at a B2B SaaS company fields 40-60 tickets daily about product features, integrations, and troubleshooting. Their internal Notion wiki is three months stale and search returns outdated answers. They add Perplexity AI to Switchy with a single API key and wire it into their support chat workspace. Now agents query live web sources and recent documentation during live chats without leaving the thread. The MCP returns synthesized answers with citations, so agents verify before pasting. This works when your product changes faster than your docs team can update internal knowledge bases. If your support needs are purely internal-policy or customer-data lookups, skip this—Perplexity searches the open web, not your private systems.

Fast competitor intel gathering for product roadmap decisions

A 3-person product team at a fintech startup runs quarterly competitive analysis to inform roadmap prioritization. They used to spend 4-5 hours per competitor manually browsing sites, reading reviews, and compiling feature matrices. With Perplexity AI in Switchy, they prompt once per competitor and get a synthesized overview of recent feature launches, pricing changes, and user sentiment from multiple sources in under two minutes. The single-tool MCP keeps the workflow simple—no scope configuration, just API key and go. This setup wins when you need broad, current web intelligence on a recurring but infrequent basis. If you're doing deep, ongoing competitive monitoring with structured data extraction, you'll want a dedicated scraping or monitoring MCP instead.

Accelerating new-hire context gathering in technical roles

A 12-person engineering team onboards 2-3 developers per quarter. New hires spend their first week reading internal docs, but also need to understand the broader ecosystem—competitor architectures, open-source alternatives, and industry best practices their codebase references. The team adds Perplexity AI to a shared Switchy onboarding workspace where new hires can ask contextual questions like 'how does Stripe handle idempotency' or 'what are current GraphQL federation patterns' and get cited, multi-source answers. This reduces Slack interruptions to senior engineers and gives new hires a self-service research layer. The single-tool simplicity means no permission overhead. If your onboarding is purely internal-systems focused, this won't help—but for roles requiring external technical context, it cuts ramp time by days.