Ai Solutions Architect

Descrizione dell’offerta di lavoro

About CleverChain

CleverChain è una azienda RegTech in fase di crescita, una piattaforma KYB premiata da Chartis Research per Best Know Your Business (KYB) e da Datos Insights per Best KYC/KYB Innovation. La nostra piattaforma cloud automatizza i processi di compliance, semplificando l’onboarding dei clienti e il monitoring del rischio per istituzioni finanziarie, fintech, provider di pagamento e tutte le organizzazioni che necessitano di verificare e gestire i propri clienti. Siamo in un momento cruciale di crescita e stiamo cercando assunzioni fondamentali per portarci avanti.

Role Overview

Abbiamo sviluppato una piattaforma di orchestrazione agentica proprietaria, il motore che alimenta tutti i nostri prodotti. Questa piattaforma coordina modelli di linguaggio (LLM), agenti, strumenti e flussi di lavoro multi-step per automatizzare complessi processi di compliance su larga scala. È una delle più ambiziose imprese di ingegneria mai intraprese, e abbiamo bisogno di qualcuno che ci aiuti a sfruttarne appieno il potenziale.

Stiamo cercando un AI Solutions Architect, chi vive e respira l’AI agentica, l’engineering di prompt e l’automazione dei flussi di lavoro. Deve aver dedicato molto tempo alla progettazione, configurazione e ottimizzazione di flussi di lavoro complessi e di sistemi di agenti, e deve capire come far lavorare in modo affidabile i LLM all’interno di livelli di orchestrazione complessi. Deve essere a suo agio nella creazione di prompt precisi e di schemi di output strutturati e nello stesso modo nella progettazione dei flussi di lavoro multi-step in cui essi si inseriscono. Criticamente, il nuovo architetto deve avere un istinto deciso per distinguere quando un passaggio richiede logica deterministica e quando richiede ragionamento guidato da LLM, e come progettare flussi di lavoro che combinino entrambi efficacemente.

Questa non è una tipica posizione di ingegneria del software. Non stiamo cercando qualcuno che scriva codice applicativo. Cerchiamo qualcuno che abbia una solida esperienza pratica nella costruzione di automazioni sofisticate e di flussi di lavoro agentici su piattaforme come n8n, Make, Zapier o simili, e che possa portare quella mentalità nella configurazione e nell’estensione della nostra piattaforma proprietaria. Deve essere quello che, davanti a un processo manuale, inizi immediatamente a mappare gli agenti, la logica decisionali, i percorsi di ramificazione e le catene di prompt che potrebbero automatizzarlo in modo end-to-end.

Avrà lavoro su due dimensioni: Core Customisations , configurare la nostra piattaforma per soddisfare le esigenze specifiche di clienti individuali; e Product Innovation , collaborare con il team per affinare e semplificare gli agenti, i prompt e i flussi di lavoro che costituiscono il nostro core. Sarà decisivo per aiutarci (e per i nostri clienti) a ottenere il massimo da una tecnologia veramente rivoluzionaria.

What You’ll Be Doing

• Configuring and customising our Agentic Orchestration Platform for customer deployments; tailoring workflows, agents, and prompt logic to meet specific client requirements.
• Designing and building multi-step agentic workflows that coordinate LLMs, tools, APIs, and decision logic into reliable, production-grade automations.
• Crafting and refining prompts across the platform; writing system prompts, defining structured output schemas, and iterating until agent behaviour is consistent and robust.
• Optimising existing workflows and agents for performance, reliability, cost‑efficiency, and accuracy; identifying bottlenecks, redundancies, and opportunities to simplify.
• Working across multiple LLM providers and models (OpenAI, Anthropic, Google, open‑source), selecting and configuring the right model for each task within a broader orchestration.
• Collaborating with the product and engineering teams to translate workflow improvements and patterns you discover in customer work back into the core platform.
• Documenting workflow architectures, prompt patterns, agent configurations, and best practices so the team can scale what works.
• Staying at the frontier ; tracking new models, new agentic frameworks, new orchestration patterns, and bringing back what's useful.

Skills & Experience

Must-Haves

• Extensive hands‑on experience building complex workflows and automations using platforms such as n8n, Make, Zapier, or similar tools. Must have real depth; multi‑step, conditional, branching, API‑integrated workflows, not just simple two‑step automations.
• Strong prompt‑engineering ability; you know how to design system prompts that produce reliable, structured, and context‑aware responses from LLMs, and you know how to debug them when they don’t.
• A deep understanding of agentic AI patterns; tool use, planning, multi‑step reasoning, agent loops, handoffs, error handling, and how to orchestrate multiple agents within a single workflow.
• Experience with structured outputs; JSON Schema, function calling, output parsing; and the practical know‑how to make models return clean, schema‑compliant responses consistently.
• Familiarity with the LLM landscape; you know the major providers and models, their strengths and trade‑offs, and you can make informed decisions about which model to use where.
• Excellent communication skills; you can explain complex workflow logic and agent behaviour to both technical and non-technical stakeholders, and you can translate customer requirements into well‑architected solutions.
• A genuine, demonstrable passion for agentic AI and workflow automation; this space moves fast and you’re already keeping up. You have opinions on orchestration frameworks, you’ve tested the latest models, and you’re excited about what’s becoming possible.

Nice-to-Haves

• Experience working in a customer‑facing or solutions role; configuring, customising, or deploying technology for clients.
• Familiarity with RegTech, compliance, KYB/KYC, or financial services workflows.
• Exposure to search‑grounded generation, RAG architectures, or retrieval‑augmented workflows.
• Basic programming ability (Python or JavaScript); enough to read code, work with APIs, and write light scripting when needed, even if you’re not a developer by trade.
• Experience with evaluation techniques for assessing LLM and agent output quality at scale.

The Kind of Person Who Thrives Here

• You're a builder. You see a problem and you immediately start thinking about the workflow that solves it. You’ve probably built automations just for fun, just to see if you could.
• You think in systems. You don’t just think about a single prompt or a single agent; you think about the full orchestration: what happens upstream, what happens downstream, where the failure modes are, and how the pieces fit together.
• You're relentlessly iterative. A workflow that “mostly works” isn’t done. You optimise, simplify, and harden until it’s production‑ready.
• You're self‑directed. You don’t need a detailed brief for every task. Give you a problem and access to the platform, and you’ll come back with a solution and three ideas for making it better.
• You're adaptable. New models drop, new capabilities emerge, best practices shift. You don’t just keep up; you enjoy the pace.
• You're customer‑minded. You understand that the workflows you configure will be used by real people to solve real problems, and you care about getting the details right for them.

What We Offer

• Work at the Heart of Our Platform: Our Agentic Orchestration Platform is the core of everything we do. You won’t be on the periphery; you’ll be one of the people who shapes how it’s configured, extended, and evolved.
• Real Problems, Real Impact: Search‑grounded generation, multi‑agent orchestration, structured output reliability, customer‑specific workflow customisation; the work is complex, varied, and consequential. What you configure ships and gets used.
• Be Part of Something Early: We’re at the stage where the right person can define how this function works; the patterns, the practices, the playbook. You won’t be slotting into a rigid process; you’ll be helping to build it.
• Direct Customer Impact: You’ll see your work in action. The workflow you configure this week will be solving a real compliance problem for a real customer next week.
• Flexible Working: Remote‑first, with core hours for alignment. We trust you to manage your time and get the work done in the way that works best for you.
• A Culture of Trust and Collaboration: Open communication, mutual respect, and a shared commitment to building something that matters. We’re a small, focused team where everyone’s voice counts.
• Growth: As we scale, this role scales with us. There’s room to grow into a leadership position within AI solutions, to shape how we deploy and optimise our platform across an expanding customer base, and to deepen your expertise at the cutting edge of applied agentic AI.

How to Apply

Send us your CV and a brief note about what caught your interest. If you have examples of workflows you’ve built, agent systems you’ve configured, prompt libraries you’ve developed, or anything else that shows how you think about orchestration and automation, we’d genuinely love to see it — but it’s not required.

Our process is straightforward: a conversation about your experience, followed by a practical discussion about real problems.