Alex Roy Rajan kept falling. On a New Year holiday in Georgia, he was taking ski lessons, and the mountain was supplying feedback with admirable consistency. His instructor watched how he leaned, noticed what was going wrong and changed the guidance. Rajan later turned the experience into a marketing lesson. The instructor had paid attention to the particular person in front of him. Anyone who has received an exquisitely personalized email about something utterly irrelevant will appreciate the distinction.
Rajan is the founder and CEO of SalesboxAI, a business he started in 2014. Much of his working life has involved the machinery that sits between a company and a possible customer. He has written software, worked on marketing automation, managed products and developed audiences. His present ambition is to make AI useful inside that machinery. The ski story offers a pleasantly undignified way into the subject: more activity means little if the response never changes.
A career on both sides of the software
His route into entrepreneurship began with electronics and communication engineering at the University of Calicut, where he studied from 1996 to 2000. He went on to software engineering roles at Tata Consultancy Services and IBM. At CDC Software, he became an engineering team lead working on Marketfirst, a marketing automation product. The commercial problems that occupy him today therefore have a technical history. Before discussing how marketers should use software, he worked on building it.
In 2007, he moved to Regalix as a product marketing manager and completed a General Management Program in IT at the Indian Institute of Management Calcutta. Product work brought market requirements, design concepts and coordination with other teams into his remit. He subsequently became vice president of product and audience development at True Influence. Engineering, product marketing and audiences are different jobs, but together they give a founder several views of the same transaction.
It would be convenient to describe each position as a deliberate rehearsal for the company that followed. Careers rarely arrive with such considerate labeling. What the sequence does establish is experience with both the making and the selling of enterprise software. A product has to function. Someone has to understand why it matters. A prospective customer has to encounter that explanation at a useful moment. Rajan’s work has crossed those boundaries.
Before the agents, a sales contest
One early chapter has a name that could have escaped from a mobile games menu: Sales Crush Saga. The sales contest app belonged to SalesBox LLC and was designed around outbound calling and motivating sales representatives. In February 2016, the company announced that it had been named a finalist in the Stevie Awards for Sales & Customer Service. Alex Roy, as Rajan is often billed professionally, appeared in the announcement as founder and CEO.
The detail is useful because it places his business before the present enthusiasm for agents. Here was a company working on sales behavior through a contest app. The interface and vocabulary would evolve considerably. The underlying territory was already recognizable: helping a commercial team do its work. A sales process contains people, incentives, waiting and repetition. Even the most industrious software has to reckon with the habits of the people using it.
A decade later, in February 2026, SalesboxAI announced a Bronze Stevie Award in the Marketing Solution category. That recognition belongs to the company and its team. For Rajan’s story, the two announcements provide dated markers rather than a substitute for an argument. They bookend a substantial change in the kind of software he was putting before customers, from a specific sales application to a broader approach to coordinating commercial work.
The avatar detour
SalesboxAI also sits beside a more unusual project in his public record. Everlife lists Rajan as its founder. Its stated ambition is a network of personal AI avatars that can extend their owners’ capabilities, preserve their knowledge and participate in work. The team includes engineering leader Charles Lobo. The proposal carries the idea of software representation beyond a sales department and into the question of what a person might delegate to a digital counterpart.
Read alongside SalesboxAI, Everlife suggests a recurring interest in software that acts on someone’s behalf. The contexts differ, and the ambition should be understood as an ambition. Still, the family resemblance is clear. A system can store information, but an assistant is expected to do something with it. Once action enters the picture, so do questions of context and responsibility. Whose priorities does the software carry into the next interaction?
His interests also reach into education. Rajan is publicly listed as an angel investor in Infusory, the company behind TutAR, which supplies augmented-reality teaching aids. It is a separate business, with separate founders. The connection adds another dimension to his technology interests: software can change how an idea is encountered, whether the audience is evaluating a business purchase or trying to understand something in a classroom.
Who is managing the assistants?
In a public demonstration of SalesboxAI, Rajan describes an arrangement with a copilot between the marketer and a group of specialized agents. The person works through the copilot; the agents perform different jobs beneath it. His demonstration includes account discovery, intent monitoring, advertising and messaging. The distinction matters to his conception of the product. A marketer should be able to direct the work without personally conducting every exchange among the software components.
The organizational idea is almost more interesting than the agent vocabulary. Adding helpers can create another management task. Someone still has to decide what the helpers are trying to accomplish and whether their activities make sense together. Rajan’s presentation tries to give that coordination a visible place. It is a product design choice with a familiar human problem behind it: having several busy colleagues is useful only if they understand the assignment.
His whitepaper on the Forrester B2B Revenue Waterfall puts buying groups at the center of the process. It considers identifying relevant accounts, finding the people involved, detecting activity, engaging them and assessing readiness for sales. The proposed role of AI changes along that route. Sometimes the job is to observe; sometimes it is to organize information or support a conversation. Treating all those tasks as one undifferentiated act of automation would miss the point of the framework.
The buyer has acquired a research assistant
Rajan’s argument also extends to the other side of the purchase. In his March 2026 discussion of B2B marketing, he explains that AI assistants now help people research vendors and compare offerings. Businesses already sell to groups whose members bring different concerns. They must also consider the software those people consult. A potential customer may encounter a summary or comparison before encountering the carefully arranged presentation a seller would prefer them to see.
That gives his interest in AI visibility a concrete setting. Product information has to be available for the research process in which a buyer actually participates. The implications run through pricing comparisons, explanations and the material a group uses to reach a decision. For a founder of marketing software, this is a change in the audience itself. The message may pass through an additional interpreter before anyone decides to speak to sales.
Buying fewer seams
By June 2026, Rajan was making an explicitly operational argument for building SalesboxAI’s lead generation infrastructure in-house. He described a business landscape assembled from data vendors, enrichment tools, outreach products and outside execution. His objection concerned control and accountability across the whole system. If one part fails, a collection of competent suppliers can still leave a company struggling to understand where the problem began.
His answer is to own more of the critical infrastructure, connecting data, verification and execution. It is a founder’s operating thesis, with a commercial product attached. Its attraction is understandable: fewer handoffs might make the system easier to direct. Its test is practical. The value would appear in the quality of the information and the usefulness of the resulting action, rather than in the number of technologies gathered under one roof.
“Personalization without relevance is just noise.”
Alex Roy Rajan
One page is a place to begin
For all the breadth of that ambition, Rajan has offered a bounded starting point. In conversation with Matt Heinz on Sales Pipeline Radio, he recommended a phased approach to adoption: begin with a subset of a website, such as a product page, observe the signals there and organize a response. The advice makes the work inspectable. A team can examine what it is seeing before allowing a larger system to act on its behalf.
In his February 2026 conversation with Brian Thomas on The Digital Executive, the horizon was broader: coordinated AI supporting revenue teams while preserving empathy, relevance and creativity. Those qualities are easy to endorse and harder to express in a workflow. They require decisions about what matters to a person in a particular situation. Rajan’s public argument keeps returning to that point even when the surrounding machinery becomes more elaborate.
The ski lesson gives the idea a human scale. A student struggles; an instructor notices; the guidance changes. Rajan’s professional project is much larger and more complicated, but that small sequence offers a useful standard for judging it. Software can discover an account, assemble information and send a message. The worthwhile question arrives just before it acts: has it understood enough to be helpful?