Savvy

An assistant that whispers what to say during a meeting

MeetingsGitHubVirtual Assistants

▲ 0 votesLaunched Aug 27, 2026

Visit Website

Daily #23Weekly #93

Savvy is a macOS meeting assistant grounded in your own documents, not the open web. Point it at a folder and it builds a versioned brief per client. In the meeting it stays quiet, then speaks up for exactly three reasons: they asked a question your brief answers, someone crossed a red line you set, or you pressed Advice. Every card cites its source. Documents, indexes and transcripts stay on your Mac; only excerpts and the audio stream leave. Apple Silicon, macOS 13+. Open source, MIT.

AI Analysis

📝 Summary

Savvy is a privacy-first macOS AI meeting assistant that operates exclusively from user-provided local documents rather than web data. Users specify a folder to automatically generate versioned client briefs. In meetings, it listens silently and intervenes only for three triggers: answering a question from the brief, detecting violation of preset red lines, or on manual 'Advice' request. Every suggestion cites its exact source document. All indexing, documents, and transcripts remain on-device (Apple Silicon, macOS 13+); only audio stream and brief excerpts are processed externally. Open-sourced under MIT license. It solves pain points of forgetting key client details, crossing professional boundaries, and compromising sensitive data privacy in meetings. Value proposition: a discreet, trustworthy local whisperer enhancing meeting outcomes with full transparency and zero cloud dependency.

📈 Market Timing

In 2025-2026, market timing is highly favorable due to rapid maturation of on-device AI (e.g., Apple Intelligence), heightened enterprise and individual privacy regulations, and sustained demand for productivity tools in remote/hybrid work. Post-AI hype, users seek practical, trustworthy local solutions over cloud-heavy apps. Economic pressures favor efficient, low-cost tools like this open-source option. Excellent Timing.

✅ Feasibility

High feasibility. Technical challenges (local document indexing, real-time audio analysis, on-device LLM inference) are well-supported by current Apple Silicon capabilities and existing open-source frameworks. Development and operation costs are low-to-medium as it's already functional and MIT open-source, enabling community contributions. Minimal supply chain or compliance risks due to fully local data handling. Strong scalability for individual professionals; team fit is ideal for solo developers or small AI teams. Key risks are macOS version compatibility and model accuracy.

🎯 Target Market

Primary segments: Tech-savvy Mac users in client-facing roles including sales professionals, consultants, lawyers, and account managers (ages 28-50). Industries: B2B services, legal, consulting, tech sales. Geographic focus: North America and Europe. Estimated TAM for AI meeting assistants ~$10B+, SAM for privacy-focused local tools ~$1B, SOM for Mac-specific apps ~$150M. Core pain points: inability to instantly recall client-specific info or boundaries during live discussions and fear of sharing sensitive docs with cloud AI. High willingness to pay for premium support or enterprise versions despite open-source core.

⚔️ Competition

Medium. Direct competitors: 1. Otter.ai (otter.ai) - transcription and notes. 2. Fireflies.ai (fireflies.ai) - AI meeting assistant with search. 3. Gong.io (gong.io) - revenue intelligence for sales calls. 4. MeetGeek (meetgeek.ai) - automated insights. 5. Limitless (limitless.ai) - wearable + AI recall. Advantages: superior on-device privacy, document-grounded briefs with source citations, minimal intervention design. Disadvantages: Mac-only, lacks broad transcription/features of cloud competitors, open-source may limit polished enterprise support and monetization compared to funded rivals.

Upgrade Pro to unlock full AI analysis