Tenth Revolution Group is looking for a Solutions Architect.
A brief about the company: it’s a giant. A global leader in IT recruitment, especially in cloud technologies. Under their wing are brands like Nigel Frank and Mason Frank. It’s amusing that a company that profits from finding people for others is itself looking for someone for a role that could already be successfully delegated to artificial intelligence today.
What pain point are they trying to solve? Let’s imagine a typical dialogue that, I’m sure, happens regularly there.
— Mark (from Sales): “We’ve almost closed a deal with a major retailer! They need a hundred of our specialists for an Azure migration project. A three-million contract!”
— Anna (Client’s CTO): “Mark, everything sounds great. But our security department wants to understand how your solution will integrate with our legacy ERP system. Show us the architecture. And also, we have strict GDPR requirements. How will you ensure data security? Please send the technical part for our tender by tomorrow.”
And that’s when Mark starts to quietly panic. He’s a salesperson, not an engineer. He can’t draw an architecture, doesn’t know all the intricacies of API and GDPR. The deal hangs in the balance. The company loses money and, even worse, reputation. They need a “translator” from business language to technical language. Someone who will sit next to Mark, charm Anna with deep knowledge, and quickly prepare an impeccable technical response for the tender.
Now imagine that instead of searching for this rare and expensive specialist, Tenth Revolution Group invests in creating a “Digital Architect.” This isn’t a robot in a suit, but an intelligent system that solves the same problem.
Here’s how it could look.
Instead of one human expert, a system is created that becomes the collective intelligence of all the company’s best architects. This is not a replacement for people, but a multiplier of their knowledge.
Step 1: Knowledge Base Creation. The company already possesses an invaluable asset – dozens and hundreds of successful projects. All technical proposals, architectural diagrams, tender responses, security and compliance documentation are uploaded into a single vector database. This becomes the “memory” of our AI.
Step 2: Integration with a Large Language Model (LLM). A powerful model like GPT-4o or Claude 3 Opus is taken and connected to this knowledge base using RAG (Retrieval-Augmented Generation) technology. Now, the AI doesn’t just fantasize, but builds its answers based on real, proven, and successful company cases.
Step 3: Interface Development for the Sales Department. Mark gets a simple tool. He uploads the client’s request into the system – be it an email, text from an RFP, or just a call transcript. And in response, he receives:
1. A draft architectural integration scheme, based on similar cases.
2. A ready-made draft technical response for the tender, taking into account the specifics of the client’s industry (retail, fintech, public sector).
3. A list of potential risks and questions that technical director Anna might ask, with ready-made answer options.
4. Links to internal documentation supporting each point.
To reduce the distrust that always arises with AI implementation, there’s no need to fire all architects. On the contrary, the remaining experts transition to a new role: they become curators and trainers of the system. They review and improve its responses, add new successful cases to the knowledge base, thereby constantly increasing the “qualification” of the digital assistant. Salesperson Mark no longer bothers engineers with minor issues, but comes to them with an 80% ready solution from the AI for final validation.
How to check if this “Digital Architect” works?
It’s elementary. Results are measured by the same metrics listed in the job description.
1. Tender win rate. Compare the percentage of won tenders for teams using the AI assistant versus those working the old way. I’m confident that the speed and depth of AI processing will make a difference.
2. Integration timelines vs plan. By analyzing dozens of past projects, the AI can provide more accurate timeline estimates, reducing the number of missed deadlines.
3. Customer satisfaction (NPS). Technical director Anna will receive her answer not “tomorrow evening,” but an hour after the call. A quick and competent response is a direct path to increased trust and client satisfaction.
4. Revenue growth (ARR). Shortening the sales cycle and increasing the percentage of won deals directly impact revenue.
Ultimately, the company gets not one employee who can get sick, go on vacation, or quit, but a scalable, continuously learning system. A system that makes the entire sales department technically savvy. And this, in my opinion, is a far more strategic investment than another line item in the organizational chart.
Источник: https://www.linkedin.com/jobs/view/4412115158/
