Company profile · 4 min read
Nessie: Shared context across AI work and agent sessions
Nessie syncs AI conversations, agent traces, meetings and notes into searchable, shareable context for people and agents.
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
Published · Updated
Nessie is a shared context layer for AI-native work. It fits a founder, researcher or team whose important decisions are scattered across ChatGPT, Claude, Codex, meetings, notes and agent sessions, and who wants agents to resume work without starting from zero.
What it does
Nessie’s current site syncs AI conversations and agent traces across ChatGPT, Claude, Gemini, Perplexity, Claude Code, Codex, Cursor and other tools, alongside meetings and notes from Granola, Read AI and Obsidian. It turns that history into searchable contexts and a living company ontology: what was decided, who owns it and how the organization’s thinking changed.
The product has both individual and team paths. A person can search and resume an old agent session. A team can share contexts, sessions and folders with permissions. Agents can use the CLI, MCP and skills to retrieve transcripts or generate context. Nessie says it is local-first by default, with optional cloud sync for teams, multi-device use and remote access; it also describes on-prem deployment for organizations that need it.
Pricing is public. The pricing page lists Plus at $28/month, Pro at $79/month, Team at $80 per seat/month and custom on-prem. Plus supports five connected accounts; Pro adds unlimited accounts and context/chat usage; Team adds administration, analytics, shared skills and ontology. That makes the key choice clear: personal memory, unlimited individual context or shared organizational memory.
Founder context and tradeoffs
The YC profile identifies Anna Zhang and Tiger Wang as founders, both Yale and former Amazon engineers. Their public backgrounds include recommendation systems, planetary-scale authentication and neuroscience/AI context. The product’s local-first and agent-access boundaries are central to the buyer decision, not decorative privacy copy.
The buyer should ask exactly what syncs, where data is stored, how third-party model processing works, how permissions propagate to agents and what a team can delete or export. A shared context layer compounds good history—but it can also compound stale or incorrect decisions if editing and source provenance are weak.
Editorial take
I would shortlist Nessie for an AI-native team already losing time to repeated context and handoffs. Start with one integration set and one recurring workflow, such as resuming engineering sessions or preparing weekly decisions. If the organization does not yet create reusable context, the system may become another archive instead of a shared memory.
Quick facts
| Field | Sourced detail |
|---|---|
| Product | Cross-tool AI history, context, ontology and agent-access layer |
| Buyers | AI-native individuals, founders and teams |
| Public pricing | Plus $28/month; Pro $79/month; Team $80/seat/month; on-prem custom |
| Integrations | AI chats, coding agents, meetings, notes, CLI and MCP listed |
| Main gate | Privacy, permissions, provenance, deletion and stale-context handling |
Sources checked
| Source | Checked |
|---|---|
| YC profile | 2026-09-19 |
| Nessie homepage | 2026-09-19 |
| Nessie pricing | 2026-09-19 |
| Nessie docs | 2026-09-19 |
Cohort context
Nessie is listed in Fall 2025. In our 2026-09-18 directory snapshot, 90 of 146 listed companies in that cohort have YC’s primary industry label B2B (61.6%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.
Public website snapshot
Observed 2026-09-19T16:15:13.152Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.
| Signal | Homepage observation |
|---|---|
| Product description metadata | Observed |
| Canonical link | Not observed in this response |
| H1 or H2 heading | Observed |
| Typed structured data | Not observed in this response |
| Docs/developer link | Observed |
| Pricing link | Observed |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |
Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.
About the author
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.
First1000 ↗ · X ↗