Lauren Rothrock, Chief Product Officer at Litify – Interview Series

Lauren Rothrock, Chief Product Officer at Litify, leads product strategy and innovation for the company as it expands its use of artificial intelligence and automation across legal workflows. She joined Litify in 2023 after serving as Vice President of Strategy & Innovation at Insight Partners, where she worked with technology companies on growth and strategic initiatives. Earlier, Rothrock spent more than six years at TravelClick, progressing through product leadership roles before becoming Vice President and General Manager of Web and Video Solutions. Her background also includes strategy and financial planning at The Leading Hotels of the World and banking and capital markets work at PwC, giving her experience spanning product development, technology strategy, finance, and enterprise software.

Litify is a legal technology company founded in 2016 that provides an AI-native platform for law firms and corporate legal departments. Built on enterprise cloud infrastructure including Salesforce, the platform brings matter and case management, documents, intake, analytics, billing, workflow automation, and AI into a unified environment. Litify has increasingly positioned its technology as a “Platform of Action,” designed not just to store legal data but to use that data to initiate and execute workflows. Its Litify Agentic Case Expert (ACE) serves as an agentic AI layer that can analyze matter context, identify risks and opportunities, and carry out multi-step tasks across areas such as intake, litigation, policy analysis, treatment tracking, and demand preparation.

Your career has taken you from banking and capital markets at PwC, through product leadership at TravelClick, to strategy and innovation at Insight Partners, and now to leading product at Litify. How have those different perspectives shaped the way you think about building and deploying AI products for highly regulated, high-stakes industries?

Starting my career as a CPA gave me a deep respect for ethics, compliance, and rigor that directly mirrors the legal world, making it very natural for me to build technology for high-stakes, regulated industries. At my core, though, my strength in product management is asking the right questions, deeply listening to client pain points, and building solutions that solve their hardest operational challenges. The advent of AI has made this work exponentially more exciting, allowing my team to tackle those high-value pain points faster and more effectively while maintaining strict governance every step of the way.

Litify describes ACE as moving beyond a traditional AI assistant toward an agent capable of understanding context and taking action. From a product and technical perspective, what separates a genuinely agentic system from a sophisticated chatbot with workflow integrations?

To put it simply, it really comes down to read-write capabilities—what can the agent read, and what can it write?

We built Litify ACE directly into our platform so its read access matches the user’s access. If a user can see a case detail, a document, or cross-case data, ACE can read it too. But we also extended that beyond Litify. Through partnerships and MCPs (Model Context Protocol), ACE connects into external accounting systems, legal knowledge bases, and third-party document repositories to get a complete picture of the context. On top of that, we layered in advanced LLM capabilities and trained the system alongside legal experts to embed legal domain intelligence directly into the experience.

Where ACE shifts from a basic chatbot to a true agent is on the write side. It doesn’t just sit in a sidebar offering advice; it can enter data, spin up new objects, update fields, and execute multi-step workflows like a member of your team. For instance, ACE can review an incoming invoice against billing guidelines and take action, track where a client is in their treatment plan and automatically send an SMS or email update, or even initiate a call to remind them of an upcoming appointment. Because ACE has full read-write capabilities natively embedded across the entire Litify ecosystem, it moves past basic chat and actually drives end-to-end legal workflows forward.

With the latest evolution of ACE, the system can reason across entire case portfolios rather than being limited to an individual matter. What new capabilities become possible when an AI agent can understand patterns, risks, and opportunities across thousands of cases simultaneously?

It’s the difference between ACE acting like a team member executing tasks on your behalf versus ACE acting as a strategic thought partner.

When an agent can analyze data across your entire portfolio—or across every matter a specific user has access to—you can start asking much deeper, higher-level questions. For instance, you can ask ACE to evaluate potential outcomes and generate case forecasts. Because so much contextual data is fed into the AI, it can draw on past performance to give you predictive insights, build comparable case sets (comps), and estimate where a new matter is likely to land based on similar historical cases.

For a firm owner or operational leader, portfolio-wide reasoning also changes how you manage resources and exposure. You can ask: “Which team members are over- or under-utilized right now, and how should we redistribute matters to balance the workload?” or “Based on our historical outcomes, who is the best-fit team or co-counsel for this new, high-stakes case?” You can also ask ACE to run portfolio-wide risk analyses—identifying where exposure lies across thousands of matters, explaining why, and surfacing recommended next steps.

Beyond strategy and forecasting, portfolio-wide context unlocks massive operational speed. A staff attorney could ask ACE to surface every matter where a client hasn’t been contacted in 30 days. Instead of just generating a passive list, ACE can draft and send personalized, context-aware check-in texts to all of those clients based on the specific stage of their case. Or, you could ask ACE to identify every active case involving a specific medical provider or vendor and automatically create follow-up tasks across that entire group at once. It moves firm management from reactive checking to proactive, portfolio-wide execution.

Enterprise AI is ultimately constrained by the context it can access. How is Litify approaching the challenge of giving ACE access to information spread across systems such as iManage, Google Drive, and SharePoint without forcing organizations to centralize or duplicate all of that data?

Most of our clients rely on Litify’s built-in document management system, Docrio. Because of that, ACE was designed to be fully, natively integrated into Docrio out of the box. When your case data and document management live in the same native environment, the depth of context and read-write actionability you get is unmatched.

That said, we know a portion of enterprise clients use external repositories like iManage, Google Drive, or SharePoint. We didn’t want to force those organizations to undergo massive data migration or duplication projects just to leverage agentic AI. Instead, we built Model Context Protocol (MCP) integrations directly with these third-party systems. This lets ACE connect into those external environments on the fly, retrieve the necessary document context, and bring it straight into the workflow.

While a native setup with Docrio gives you the deepest, tightest read-write capabilities, using our MCP architecture for external repositories still delivers a seamless, highly powerful experience—and it’s head and shoulders above traditional, disconnected document search.

Permissions become particularly complicated when an AI agent is operating across multiple repositories and applications. How do you ensure that ACE inherits and respects the access controls already established for individual users, documents, and matters?

Litify is built natively on Salesforce, and much of our platform relies on Salesforce APIs—so governance, security, and granular permissioning are in our DNA.

We built Litify ACE to mirror and enforce every specific, detailed permission and security model configured in Salesforce. When a user interacts with ACE, the agent dynamically inherits that specific user’s exact permission boundaries. If a user doesn’t have explicit permission to see a specific matter, field, or document, ACE simply cannot read, process, or act on that data for them. This is an area where Litify is absolute best-in-class. Enterprise legal teams can’t afford data leaks across matters or roles, and our native inheritance of enterprise permissions gives firms peace of mind.

As agents progress from generating answers to executing multi-step workflows, the consequences of an incorrect decision become more significant. How do you determine which actions ACE can take autonomously and where human review or approval should remain mandatory?

The short answer is that Litify puts this control directly in the hands of our clients—they get to decide exactly where there’s a human in the loop and where an agent can operate autonomously.

Beyond a simple binary choice between full automation and human review, our clients can configure specific guardrails, such as financial or risk thresholds. For example, we have an agent that handles invoice adjustments when line items are disputed. A firm can choose whether that agent simply suggests a recommendation to a human billing manager or executes the adjustment directly. But even if they enable autonomous execution, they can set a dollar threshold—saying the agent can automatically adjust disputed items under $500, but anything over that requires human sign-off. It’s all about giving firms the flexibility to balance speed and productivity with their own risk tolerance.

Trust is especially important in legal environments, where professionals need to understand the basis for an AI-generated recommendation. What types of citations, evidence trails, action histories, or other verification mechanisms do you believe enterprise AI systems need to provide?

Unless a fact is completely ubiquitous, every piece of data used by AI must be clearly cited. ACE provides exact inline links and citations down to the specific field, document, or page number, as well as direct links to official legal authorities when referencing case law. For transparency around actions taken, we integrate seamlessly into Litify’s core Activity Timeline, which logs every document saved, task updated, or message sent or received across a matter. We’ve added a dedicated ACE agent persona to that timeline, so users can see a complete, timestamped audit trail of every autonomous action taken by the AI right alongside the work of their human team members.

One of the promises of agentic AI is that users should not have to become expert prompt engineers. How do you design an agent that can infer what needs to happen from the underlying matter, workflow, and organizational context rather than requiring a user to explicitly instruct it at every step?

We couldn’t agree more—legal professionals shouldn’t have to become prompt engineers to get value out of AI.

Because Litify is fundamentally a workflow engine, ACE natively understands the entire lifecycle of a matter—including the case type, estimated value, current stage, assigned team members, recent communications, and underlying documents. It uses that rich context to make intelligent, localized decisions automatically, rather than relying on a user to spoon-feed it background details in a chat box.

To make interacting with ACE effortless, we built a Skills Library. ACE ships out of the box with preconfigured “skills”—which are essentially expert-crafted, complex prompts tailored to specific legal practice areas, stages of a case, and user personas. Users don’t have to write prompts from scratch; ACE dynamically surfaces the right skill at the right time based on where they are in the workflow. Furthermore, firm admins can customize this library to standardize best practices across the organization, and individual power users can save their own successful prompts into the library for future use.

How are Litify customers measuring whether agentic AI is actually delivering value? Beyond time saved, are you seeing organizations evaluate metrics such as case throughput, consistency, risk identification, response times, or the quality of outcomes?

For our clients—some of the fastest-scaling law firms in the industry—the ultimate metric comes down to reducing time on desk and increasing throughput. Because Litify ACE automates workflows to drive matters forward, firms can scale revenue faster without endlessly expanding headcount.

Beyond overall throughput, we see value measured in a few distinct ways:

• Direct Bottom-Line Impact: For example, we’ve heard several times that ACE uncovered an overlooked case insight so valuable that it paid for an entire year of the solution in one move.

• Radical Time Savings: For example, document drafting that once took 1.5 hours now takes 10 minutes for one of our firms in Florida.

• Better Client Touchpoints: Firms are resolving more cases with the same staff while significantly improving client communication. Instead of spending hours stuck in manual administrative tasks, legal teams are staying in closer contact with clients throughout their matters, driving higher overall satisfaction and case momentum.

Looking beyond legal technology, do you see the transition from systems of record to AI-driven systems of action becoming a broader enterprise software shift? What do you think today’s enterprise applications will look like once autonomous agents become a standard layer across business workflows?

At Litify, we’ve already seen 10-20% of our customer base adopt our AI solutions, and that growth happened remarkably fast. While our products are, of course, best-in-class, I think this shows how hungry clients are for native AI built directly into their core platform of action.

Why are clients demanding native AI in their platform of action? It comes down to two big reasons: 1) Point solution fatigue. Organizations are struggling with standalone AI point solutions. Users resist adopting tools that require separate logins, isolated interfaces, and fragmented workflows. AI only delivers value when it lives directly inside the platform where daily work already happens. 2) More importantly, AI is most effective when it has full read-write access to the core workflow engine. A system of record can only truly transform into a system of action if the foundational data set and workflows are already built into the platform. AI is an intelligence layer that must run across existing operations, not instead of them. Autonomous agents require deep matter context, historical data, and structured operational alignment to execute multi-step agentic workflows end-to-end. Without direct access to drive work forward, AI remains a novelty rather than a true force multiplier. That’s why this transition has been so quick and exciting for our clients: the operational foundation was already in place, ready to be activated.

Building native, context-aware automation takes immense resources. While organizations often explore building custom agent layers using raw LLM APIs, the token costs, maintenance overhead, and engineering complexity quickly make it cost-prohibitive. Enterprises want one unified platform that delivers best-in-market AI-native to their existing operations—and that shift is going to redefine enterprise software across every industry.

Thank you for the great interview, readers who wish to learn more should visit Litify.