Sentra.app has raised a $5 million seed round to develop what it calls an “enterprise general intelligence” layer built around organizational memory: a persistent system designed to capture, connect, and continuously surface the context behind a company’s conversations, decisions, and work. The round was co-led by a16z speedrun and Together Fund, with participation from Parable, Precursor Ventures, Inovia, Backwards Capital, Antigravity Capital, and a group of angel investors including Gokul Rajaram, Siqi Chen, and former executives from Microsoft, Dropbox, Salesforce, and Slack.
The company is positioning its product as a response to a familiar enterprise problem: organizations generate an ever-expanding volume of knowledge, but much of it disappears quickly, becomes siloed across teams and tools, or loses its decision-making context over time. Sentra’s pitch is that “memory” shouldn’t be a static archive. Instead, it should function as an active layer that unifies data, conversations, and timelines, then uses that foundation to help teams align, collaborate, and make decisions with shared context.
Sentra says it is already running a paid proof-of-concept with SoftBank, which it frames as early validation that large, complex enterprises are looking for new ways to preserve institutional knowledge and prevent misalignment from compounding across global operations. The company’s approach emphasizes building a continuous “timeline” of decisions and commitments, learning an organization’s priorities, and identifying potential divergence before it spreads, with the broader goal of improving clarity and coordination at scale.
Product-wise, Sentra describes an organizational memory system that connects the “who, what, when, and why” behind key choices, turning fragmented activity into a coherent narrative that can be revisited and reused. The platform is presented as a persistent knowledge layer that supports real-time alignment and draws on historical context, aiming to serve as a single source of truth for what was decided, what was committed, and the context that led to those outcomes.
Beyond knowledge capture, Sentra highlights workflow automation as a practical on-ramp for adoption. The company lists use cases such as status reporting, meeting decision capture, actionable reminders, and just-in-time onboarding, delivered via modular apps. The underlying bet is that organizations will adopt a “memory” system faster if it immediately reduces administrative overhead while simultaneously building a durable record of institutional decision logic.
Sentra’s founding team includes CEO Jae Gwan Park and CTO Andrey Starenky, alongside Al Rey and Ashwin Gopinath, a former MIT professor. The company draws on Gopinath’s research on Reflexion, which explored agent self-improvement through feedback. Sentra argues that similar principles can be applied to organizational memory, shifting from static knowledge bases to systems that learn what matters through usage, retrieve the right context at the right time, and evolve as the organization changes.
Strategically, Sentra is aligning itself with the idea that the next wave of enterprise AI is less about replacing people with models and more about strengthening continuity of thought across teams. In that framing, AI becomes a layer that preserves context, intelligently propagates information, and accelerates decision velocity by making alignment easier to achieve and maintain. The company also emphasizes enterprise-grade security as part of its go-to-market message, reflecting the reality that memory systems require deep access to sensitive internal data and communications to be useful.
If Sentra can deliver on its promise, it would sit in an emerging category between knowledge management, collaboration tooling, and agentic automation. The company is effectively arguing that enterprises need something more foundational than another document repository or chat-based assistant: a durable, reasoning-capable memory that turns ephemeral work into a living system of record, while also making day-to-day execution faster and less error-prone.
KEY QUOTES
“Modern companies generate more knowledge than they can retain or use. Every conversation, decision, and document adds to a growing ocean of context that disappears moments later or stays siloed. Sentra proactively creates a living company memory that learns, reasons, and reflects alongside them.”
Jae Park, CEO, Sentra
“Sentra acts as the brain of your business, allowing real-time, true understanding of what your teams do. It optimizes organizational thinking, enabling effective leadership at scale where alignment is essential and clarity compounds.”
Andrey Starenky, CTO, Sentra
“In Reflexion (NeurIPS 2023), we showed how AI agents could learn by reflecting on their own reasoning. At Sentra, we’ve applied this principle to organizational memory itself, building systems that don’t just store information, but learn what matters based on use, surface the right context at the right time, and evolve as the organization does. This is how we move from static knowledge bases to true Enterprise General Intelligence.”
Ashwin Gopinath, Co-founder, Sentra
