Entelligence is tackling engineering inefficiency with AI 🎛

CV Deep Dive

Today, we’re talking with Aiswarya Sankar, Founder and CEO of Entelligence.ai.

Entelligence is tackling the inefficiencies that come with large engineering teams, focusing on making the time engineers spend outside of writing code—like onboarding, planning, and code reviews—far more efficient. With a background at Uber and a passion for solving real-world engineering challenges, Aiswarya and her team are developing tools to help engineers quickly navigate complex systems, reducing the time it takes to become productive.

Since launching, Entelligence has attracted open-source maintainers and large companies alike, helping them streamline onboarding, automate workflows, and unify the context that engineers need to do their jobs better. By pulling together everything from pull requests to documentation into a single platform, they’re working to solve the biggest problems teams face when scaling.

In this conversation, Aiswarya talks about building Entelligence, the unique challenges of creating tools for large engineering teams, and what’s next for the platform as they push forward.Let’s dive in ⚡️

Read time: 8 mins

Our Chat with Aiswarya 💬

Aiswarya - welcome to Cerebral Valley! First off, give us a bit about your background and what led you to found Entelligence? 

Thanks!  I'm Aiswarya, founder and CEO of Entelligence AI. My journey in tech began doing EECS at UC Berkeley, where I specialized in computer vision and medical imaging. 

After Berkeley, I joined Uber, where I had the opportunity to deploy the first NLP deep learning model for the Uber Freight support team, a project that significantly improved operational efficiency.  This pushed me to pursue a Master's in NLP at Stanford while continuing my work. I founded an NLP reading group through the ML Collective and worked on publishing a paper on multi-document summarization.

This led me to found my first startup, where I built our first RAG systems and multi-document summarization models from scratch prior to GPT even being launched.  Extremely exciting, but the market wasn’t yet ready for it.  Throughout this process, I noticed that at big companies like Uber and Google, engineers were only spending 20-30% of their time writing code. The rest was overhead—meetings, mentorship, talking to other teams. It wasn’t as fun or innovative as I wanted. So I thought, "How can we solve that?"

In large tech orgs, it can take up to four years to become a “domain expert,” which is the time to get an entire bachelor’s degree just to understand the engineering stack and codebases inside and out. It really should not take that long. So with Entelligence, I wanted to build an AI system that brings you that engineering system awareness from day 1.

We built a system that focuses on two main things: context awareness and planning. We built a version at Uber, won a hackathon there and had several teams across the organization reach out to want to use it. That’s when I knew this could be more than just an internal dev tool. So I quit, went all in, and we recently raised our seed round. 

Give us a top level overview of Entelligence - how would you describe the startup to those who are maybe less familiar with you? 

At Entelligence, we have a few main goals. The first is using our platform to remove knowledge silos and streamline overhead for engineering teams.  Through this we’re able to help engineers with 80% of their time when they’re not actively writing code. This includes everything from figuring out how to build things, dealing with dependencies, searching, doing code reviews, mentorship, and more. This becomes a really big issue when context at these large companies gets complex, historically and across teams.

We’ve built two core APIs that power our entire platform. The first is planning, which helps you figure out how to solve an issue based on everything available to you. The second is context—so being able to understand large codebases. These are the same challenges that a lot of leaders in this space are working on, and no one has fully solved it yet because it really cuts to the core of what software engineering is.

We’ve been working on those two pillars, and they power the rest of our system. One of the main use cases so far has been onboarding. We synthesize large codebases, break them down into steps, and help answer questions along the way. We also respond to open issues and review pull requests. Our main distribution channel has been partnering with open source maintainers who want to automate the process of onboarding, understanding, reviewing, and giving feedback. 

Our focus has really been on pulling all the context together to power the various products engineers need.

Talk us through which use-cases for Entelligence have worked best so far, or have had the biggest impact on the developer experience.

At Entelligence, we’re building artificial engineering intelligence that understands your entire engineering stack, helps wherever you need it across the entire SDLC.  This means we’ve built our system as a context and reasoning platform that powers use cases across the engineering workflow.  We’re able to answer questions on Slack, within VSCode, in our web UI and help engineers from onboarding to code reviews.  Regarding which use cases have had the biggest impact on the developer experience, I’d say different sized companies benefit different features more.  Here are a few examples - 

3rd party integrations 

Unlike every other VSCode extension, copilot, cursor, codeium etc, Entelligence allows you to pull in any other 3rd party OSS or private repo into context while you’re building.  This allows you to do entirely seamless integrations since Entelligence has access both to your codebase and full context of the codebase you’re integrating.  You simply have to highlight or reference the files where you want to do the integration in your codebase, choose the repo you’re integrating with and Entelligence can embed the integration within your code.  Check it out below! 

Chat with your engineering stack

Another very popular use case is using our engine to power your “chat with your docs” that actually covers all important knowledge sources including live updates in your codebase, new issues, PRs, and conversations within your Discord.  Our docs walk you through a streamlined process for configuring all your knowledge sources for the integration.  Admins then get access to a detailed analytics dashboard across all user questions, gaps in the current knowledge base and top trends across user activity. 

Pull Request Reviews and Team Insights

We’ve yet to talk to an engineer who finds reviewing pull requests to be their favorite part of the engineering process. Entelligence uses the reasoning engine to power pull request reviews with full codebase context, awareness of your codebase best practices and other recent PRs. Entelligence has reviewed thousands of PRs to date including on the fastest growing OSS repos such as Based Hardware. In addition to reviewing each individual PR, we also provide aggregate stats, reports and search across PRs making it extremely seamless to empower your team to merge and deploy with top quality.  Here’s an example of our insights dashboard - 

There’s a lot of excitement around AI agents today - how are you factoring that into Entelligence’s product or roadmap? 

We’re taking a different approach by recognizing that engineers are more than just coders. There's a lot more that goes into the job than just writing code. So instead of just trying to build coding bots, we’ve created a suite of agent templates for things like migrations, codebase onboarding, and pull request reviews.

We plug our endpoints into these agent frameworks so that teams can customize them to their specific needs. Take onboarding, for example—it’s a huge issue for large tech teams where getting an engineer up to speed can take six months or more. Keeping docs updated and having a senior engineer walk every new person through the process is a pain. 

So, what we’ve done is build agents that streamline this whole process. You plug in all your existing resources, provide a high-level list of key points, and the agent finds the most up-to-date info across your docs and resources. It then walks the user through the onboarding process. We’ve found this valuable not just for onboarding but also for tasks like migrations, where teams need to coach everyone through new workflows. 

At the end of the process, you can set it up to post updates to Slack or Discord or get an email on how things are going. Essentially, we’re looking at those higher-level engineering processes that require more organizational awareness and building agentic workflows around them.

How do you plan on Entelligence progressing over the next 6-12 months? Anything specific on your roadmap that new or existing customers should be excited for? 

We focus on helping engineers with tasks that go beyond just autocompleting or writing code. For example, our mentorship recap feature has been really popular with team leads and managers. It provides detailed insights into what each person has accomplished, including code changes and pull requests, all summarized in a clear, time-series-based format.

We’re also honing in on customizable agents for engineers, helping them plan out tickets and giving them deeper insights into their work. We’ve launched a new platform and interface that can be used in VS Code, the UI, Slack, or wherever support is needed most. 

Finally, we’ll be adding increased support for Entelligence artifacts! Full project and feature planning are soon to come within this interface - 

Lastly, tell us a little bit about the team and culture at Entelligence. How big is the company now, and what do you look for in prospective team members that are joining?

We’re a team of six right now, all engineers and a designer. We’re very focused on building solutions rooted in the real problems we see companies and organizations facing every day. When hiring, we focus on people who don’t see problems as roadblocks but as opportunities to do something innovative and exciting.  The palpable excitement brilliant engineers bring when facing the hardest challenges is probably the single most important thing we look for.

We also encourage everyone on the team to be critical about their own workflows. We use our product constantly to expedite our engineering.  When we realized we were writing the same endpoint scaffolding over and over, we built out our entire multi step reasoning agent to entirely automate that process for us. It’s all about making the engineering experience smoother and more efficient.

We’re actively hiring more stellar research engineers and product driven engineers! If you, like me, are driven to build artificial engineering intelligence that actually reduces engineering overhead for real engineering teams please reach out [email protected]! 

Anything else you want people to know about the work you’re doing at Entelligence? 

At Entelligence, we really want to figure out and solve what engineering intelligence actually looks like. And that’s more than just writing code—it’s about mentorship, collaboration, and reviews too. So, we're building the key technical pieces to handle those challenges and creating a platform that can scale for big engineering teams.

Just to summarize as a take home message, the two biggest things we’re focused on are planning and awareness. Large codebases can be a mess to navigate, and if you’ve got someone who’s been at a company for years, they’re obviously going to know way more than someone new. That context gives them an edge when solving problems. What we’re trying to do is give everyone that same level of insight right away, without needing years of experience at the company.

Conclusion

To stay up to date on the latest with Entelligence, learn more about them here.

Read our past few Deep Dives below:

• Our chat with Datastax's Chief Product Officer, Ed Anuff ❇️

• TensorStax is building Autonomous AI for MLOps 🌐

• Dashworks, your AI-powered internal search tool 🔍

• Delphi - the verified digital cloning platform 👥

• CodeGPT is intertwining human language with code 🌏

If you would like us to ‘Deep Dive’ a founder, team or product launch, please reply to this email ([email protected]) or DM us on Twitter or LinkedIn.

Join Slack | All Events | Jobs