DevRev was founded in 2020 with a mission to ‘to unlock team potential through human-AI collaboration’ and transform how businesses organise, search and collaborate by breaking down barriers between departments.
In 2022, it announced an AI platform, OneCRM, that connects enterprises’ support, product, sales and engineering teams and customers, before, in 2024, operationalising GenAI and natural language interactions with the launch of the AgentOS platform for its customer service, product management and software engineering apps.
This provides one-click data migration from legacy systems into a data layer unifying all product, customer and operational data; lightweight AI agents that analyse structured and unstructured data, including discussions (e.g. email and live chat) for search, analytics and workflow automation; and a conversational AI interface.
Last September, it took another big step towards achieving its goal with the launch of Computer by DevRev, an integrated AI teammate that listens, acts and understands context, remembers information and reasons across an organisation’s data in real time.
The conversational AI platform unifies all enterprise data (structured and unstructured) so that AI can use a business’s own data, as opposed to public data, to answer questions such as ‘which customers are still waiting on a support response?’ and to empower AI agents to complete more tasks themselves.
Now publicly available, Computer addresses the inefficiencies of modern work which for too long has been bedevilled by siloed information, disjointed workflows and scattered tools built for departments and not whole organisations to the detriment of team collaboration, knowledge-sharing, productivity and agility.
By liberating data from silos, including structured data, like CRM and product backlogs, and unstructured data, like docs and chats, Computer gives AI the full context it needs ‘to perform like a teammate’, from surfacing lost documents, summarising long threads and preparing meeting briefs to automating and carrying out tasks. Critically, it does this while keeping systems of record updated and in-sync.
DevRev calls this Team Intelligence. But what might that mean in practice?
The example DevRev uses is how Computer might respond to a bug affecting a business’s three largest customers: ‘A typical AI tool might generate a report about it. Computer goes further: it creates the ticket and assigns it to the right support expert. It alerts the product owner, so priorities shift before customers churn. It drafts the customer comms, all while keeping Salesforce and Jira perfectly in sync and complying with data permissions. That’s the difference between a chatbot that scrapes data and a true AI teammate that has longterm memory, reasoning skills and the power to act.’
Conversational computer
DevRev views the emerging era of the ‘conversational computer’ as the third major shift in human-computer interaction, following ‘point, click and scroll’ in the 80s and ‘swipe and tap’ in the 00s.
Not surprisingly, many other businesses are active in this area. However, DevRev is confident the patented technologies it has been developing since 2021, before the emergence of LLMs and agentic AI, give it an advantage over rivals that have not invested in the foundational services it has been working on for the last four or five years.
These are:
1. Computer Memory, a fully integrated data core based on a data warehouse and proprietary knowledge graph that turns enterprise data into a living network that maps complex relationships between teams, customers and products; and
2. Computer Airsync, a bidirectional, realtime synchronisation engine that brings data into this intelligent network, while preserving data access permissions, context and compliance requirements.
Michelin analogy
These technologies are the reason why when DevRev CTO Ahmed Bashir briefed Technology Reseller about Computer he began by talking about the Michelin Guide to the best hotels and restaurants.
“A couple of days ago I was thinking about the story of Michelin, the restaurant review book. Michelin’s really a tyre company; they are category leaders in tyres and have been from the very onset of automobiles, and the reason for that is because they had radial tyre and vulcanised rubber technology at the precise time that automobiles needed it,” he explained.
“Automobiles then weren’t limited by the power of their engines but by their tyres. Michelin had already done tremendous work trying to create the best pneumatic tires for bicycles and because they had the right radial tyre technology, even though they had built it for bicycles, they were able to apply it to automobiles. That’s how they became category leaders. It’s also the reason they got into restaurant reviews; they wanted people to travel and eat outside their homes.”
Michelin’s story is analogous to DevRev’s because technologies the company created for one purpose, to break down the barriers between developers and growth teams, have the potential to be of even greater value for agentic AI, not envisaged when they were originally developed.
“When I joined DevRev in 2021 we had the idea to explore the intersection between AI and the enterprise, so we got AI from the very beginning. But in 2021, even though the world had already introduced the concept of Transformers, the concept of LLMs and conversational search, the way we see AI today didn’t exist.
“We had already patented our first vector database, in 2021 or 2022, and unifying information and being able to help people work better together through the power of AI was something we were already building technologies around. It required us to invest tremendous time on workflows, because we needed to be able to synchronise data from disparate systems and we needed to bring in tremendous amounts of data, or on some days very little data, and be as reliable moving data that took two hours to move as moving data that took two milliseconds to move.”
From the very beginning, DevRev invested in an identity engine; an authorisation layer; a flexible data layer, enabling workflows to leverage and synchronise information from diverse databases and SaaS tools; and the ability semantically to interpret data, all of which have become foundational services for the agentic revolution we’re now in.
“The world of agentic hadn’t been introduced to the world then but the concepts had and those foundational services, which we had worked on since 2021, and our eagerness to explore that intersection between AI and enterprise meant that in 2024 we were in the perfect position to really develop agentic tools, an entire Agent Studio (for building specialised agents) and to define what we call Computer, the product we now have in market.”
From vitamin to painkiller
Bashir describes the launch of the first LLMs in 2023 as ‘transformative’ for people’s understanding of the value of unifying information which, to use his words, went from being a vitamin to a painkiller.
“When we talked about unifying information prior to 2023, people felt it was desirable but not necessary to be able to organise information so that for example your customer success team and your customer sales team or customer experience team could work from a single tool; so that the person who’s fixing the bug in the back office and the person who’s connected to the customer in the front office are able to see the same quality of information across various dimensions and various layers.
“That seemed like an awfully nice-tohave feature then, but it wasn’t a pain killer. After 2023, people felt that the ability to harness the data so that facts could ground conversational answers and subsequent actions became highly desirable. That was the major shift.”
Flexible schema
He adds that the work DevRev was doing then gives it an advantage over competitors who generally don’t have a flexible, versioned schema engine that he says is needed to unify structured and unstructured data.
“Computer’s memory is fronted by a data layer that is SQL-compliant, and to do that you need a back-end component that is a fully versioned schema that you can run an entire database operation on. Most of our competitors are not building a schema layer, nor are they building an SQL engine, in which case the information received conversationally is not going to be grounded in something highly declarative and provable like SQL. Because you don’t have that grounded fact you cannot just transition from a conversational experience to a more traditional experience, like a dashboard or a widget, that you might share in a board meeting or put on a monitor and share with the rest of your organisation because it’s grounded in something factual and declarative. We’ve been able to ground everything we’ve done because we have this schema engine, this data layer. That’s unique in the market,” he said.
Another competitive advantage, Bashir suggests, is the way in which DevRev brings in information.
“It’s not just that we’re bringing in structured and unstructured information, which is highly differentiated in the market. We’re also bringing in access permissions. We’re bringing in the schema of the product,” he said.
This is critical: a) because in the agentic world nobody’s going to recreate the entire access provisioning layer that already exists in systems of record, which you therefore cannot afford to lose during synchronisation – “We preserve that information; most of our competitors don’t,” said Bashir; and b) because DevRev brings in both the record and the underlying definition of the schema.
“We’re not bringing in data in a flattened, lossy format and saying ‘We want to be able to observe this record and be able to understand how to answer questions with this exact information’. We bring the record in, but we also bring in the underlying definition of the schema. That’s important because you don’t just want to know the city is Carmel and the state is Indiana, you need to understand that there’s a field dependency between those two attributes and you need to understand what the other drop-down values are and not just make assumptions on what they are or simply ignore what they are.
You need awareness around that so that if somebody wants to get an action answer and act on that information, they know what values are permissible. You might say ‘I want to update the city from Carmel to Birlingham’ and it could say ‘That’s not an option, but you could change it to Birmingham’. ‘Okay, fine, let me do that’. The only way to allow for that is to know what the other options are. And you only know that when you have a deep understanding of the schematics that are powering the record as it stands today.”
Why AI projects fail These capabilities, he suggests, or rather the lack of them, is one reason why so many AI project fail. “There are people in every business, in every department who are eager to on a support portal, wherever it needs to manifest. It’s multilingual. It has an entire Marketplace that allows you to synchronise information using Airsync.
You can see exactly what’s going on, every workflow, every agentic skill, every operation. You can see the sessions you can administer. You can check what roles, responsibilities, groups exist, what has been brought in from different systems. How do you interact with this data? How do you reconcile and cleanse it? A very sophisticated platform of services must exist for you to feel comfortable deploying in production.”
What’s certain is that despite deployment challenges nobody is immune from agentic AI. Mid-market enterprises have a clear mandate to optimise business processes and even small and medium sized businesses want to substitute human labour for GPU labour. “Agents are really workflows that have the repeatability of a micro service but the sensibility of a human and that combination is mighty powerful. Small companies want to act big by having these AI workers.
The joke goes ‘I have a staff of 350 engineers, of which 35 are human’. It’s the new normal for a small business. Then, of course, established enterprises want to build operational efficiencies, because that’s what’s going to help their bottom line. We’ve got hundreds of paying customers and most of them right now are mid-market enterprises.
And we’ve got a lot of smaller companies that are exploring our product; we like to have that free tier that allows them to get involved and pick their tool of choice, and as they grow into the product, there’s obviously going to be commercial opportunities.” Bashir believes business process optimisation, especially in those areas DevRev has been addressing since 2001, is where the money will be spent.
“We’ve talked to analysts from Gartner and many others and they confirm that even though there’s a lot of dialogue around how AI can really help growth for business, what’s happening is that people are spending for business process optimisation and you can optimise business processes in any and all departments. “For us, customer experience, customer support, customer success, enterprise sales and general enterprise-level search are interesting problems because they have not seen much automation.
People continue to expend lots of energy on finding information, which is why Team learn AI. And there are leaders in every business, who are eager to allow people to see what they can do, even if it doesn’t get to the finish line. I think that eagerness, the novelty and momentum of a new technology and, I would say, the desperation of business leaders to describe interesting things that are happening within their four walls have generated investment in a lot of ideas, but not necessarily enough investment to get them to the finish line.
That’s one reason why projects fail. “The other, more practical, reason is it’s not as easy as to build business logic on top of a frontier model as it looks, because you still need a data reconciliation layer, a synchronisation layer; you need a data cleansing layer; you need a way to handle data redundancy, because if you’re bringing information from disparate sources there will be data redundancy; you need an access provisioning layer; you need an orchestration layer; you need an auditability layer – you need the ability to make sure that performance is not regressing but also to see who is spending money on the technology, who is using up all the tokens, who is going to be charged back; and finally, you need to understand the application.
You have to understand the needs of an enterprise sales team versus a traditional sales search use case versus a customer success or customer support or even developer productivity use case, and because those other layers simply aren’t seeing investment, these projects fail,” he said. “That’s where companies like DevRev with Computer come in.
You need to be able to handle the data layer, the provisioning layer, the orchestration layer, the agentic layer, the application layer, and we do that through our Marketplace (of pre-built integrations with external systems), through Computer, through the surfaces we’ve created. Our product is manifesting on your desktop, but also on your website (if you need it to), on customer success, on a help portal Intelligence is such a huge opportunity.
Customer interactions, whether on the sales side or the support and success side or to understand the temperature of a specific account, are pretty high-touch today, and so naturally these happen to be layers where the business process optimisation function seems to be quite engaged.
“I’ll also say that there is a second factor here, which is that people want to make sure they are able to validate the savings, the optimisation they’ve made through AI or whatever product they’ve adopted. For that reason, KPIs are important. And it just so happens that when it comes to customer satisfaction, the NPS score happens to be a pretty robust and broadly adopted KPI. For that reason alone, customer interactions are really interesting areas for early adoption of agentic products like Computer.”
Agentic AI benchmarks One of the challenges DevRev faces is that everyone is claiming to have industry leading agentic AI, which Bashir says is creating the need for benchmarks and leaderboards like the TPC benchmark introduced for databases in the late 80s which enabled people to see very clearly how databases fared against each other in terms of performance, personalisation and accuracy.
“I think we’re going to have to build those same leaderboards and benchmarks here so people can distinguish the leaders from the pretenders,” he said. In the meantime, Computer by DevRev will continue to evolve.
“On the technology side, I think our product’s going to become increasingly collaborative and increasingly multiplayer. The surfaces on which we operate – the desktop, the web and the mobile – will refine themselves. Our product will start to become more voice-driven. And the sophistication of the Agent Studio will allow you to upskill Computer and deploy to your team, to your organisation or just to yourself agents that have been upskilled for specific brands, for specific regions or for specific channels. Our ability to drive that and to evaluate it and version it and deploy it and roll it back is going to get more and more sophisticated as well.
“The Agent Studio, the multiplayer experience and the surfaces will continue to refine themselves. The product will become more collaborative, it will become more personalised and the benchmarks will start to reveal where companies have distinguished themselves in terms of accuracy, speed and personalisation.”
Be First to Comment