AI Engineer (Backend & Agents)

About Thinkbay AI

Thinkbay AI allows enterprises to easily build and deploy AI agents that do real work inside a business: work that no humans should do in the age of AI. We build agents to automate daily workflows across Sales, Marketing, Finance and HR functions. Agents deployed on the Thinkbay platform are not chat agents, but background agents that understand the business context, automate work, with human oversight. Our belief is simple: AI agents should adapt to how you work, not the other way around.

Under the hood is our own agent platform. It is a multi-tenant, multi-product system that runs agents for many businesses at once. On top of it we ship products:

• Process automation: agents that run business processes end to end across SAP, Zoho, Tally, Salesforce, Microsoft 365, Google Workspace and more.
• Finance operations: agents that take an invoice, sales enquiries and agreement from the mailbox or WhatsApp to an ERP entry, with human approval where it matters.
• Marketing intelligence: agents that analyse Meta and Google Ads performance and tell a team what to optimise.
• Sales coaching: AI role-play that trains sales teams on realistic conversations.

Teams talk to their agents in plain language over WhatsApp, Telegram, Slack or Teams. We serve real estate, e-commerce and any business with work that repeats. Thinkbay is headquartered in Dubai, UAE (thinkbay.ai).

The role
We are hiring two entry-level AI Engineers (Backend & Agents) to build the engine that every Thinkbay product runs on.

This is not a role where you wrap a chat box around an API. You will own agents that act in production for paying businesses: they call tools, read documents, move money-adjacent data and must finish what they start. You will decide how an agent plans, which model it uses, what it remembers, when it asks a human, and how it recovers when a step fails.

You will work directly with the founding engineers, ship weekly, and own whole systems rather than tickets. The platform is young enough that your decisions will shape it for years, and real enough that they reach customers the same week.

What you'll do

• Build business products on our agent infrastructure. Create AI products for finance, accounting, marketing, HR, supply chain and other business functions, each solving a real use case. This is our main goal and the heart of this role.
• Own the agent platform. Alongside building products, maintain and scale the agent platform that powers them.
• Build agentic applications. Design agents and multi-agent workflows that plan, use tools and complete real business tasks reliably.
• Make tool calling reliable. Design the tools agents use and make sure agents handle errors gracefully.
• Build retrieval (RAG). Give agents the right knowledge and context from business data and documents.
• Work across LLM providers. Integrate and evaluate models from multiple providers, and pick the right one for each job.
• Optimise performance. Improve the cost, speed and quality of our agents, and measure the impact.
• Integrate with third-party software. Connect agents to the business tools our customers already use.
• Go multi-channel. Let users work with their agents across messaging and collaboration platforms.
• Run it in production. Monitor, debug and improve live systems, and ship with confidence.

Must-have skills

• Data structures and algorithms. Strong fundamentals and sound problem-solving.
• Python. Solid production experience with Python and backend development.
• LLM APIs. Hands-on experience building with LLM APIs.
• AI backend engineering. Experience building backend systems for AI-powered applications.
• Agentic orchestration. A good understanding of how to build and orchestrate agents that use tools across multiple steps.
• 0-3 years of work experience

Nice to have

• Experience with vector databases and retrieval systems.
• Familiarity with agent frameworks and protocols such as MCP.
• Experience evaluating and testing LLM systems.
• Experience integrating messaging platforms or business software.
• Familiarity with cloud, containers and CI/CD.
• Experience building multi-tenant SaaS products.
• Open-source work or side projects involving AI agents.

Why join Thinkbay

• Real agents, real stakes. Your work runs for businesses every day. You will see it used, and you will see it fail, which is how you get good.
• Ownership from day one. You will own whole systems: an integration, a product's agents, a part of the platform.
• The frontier, in production. You will work with the newest models and agent techniques as they ship, and turn them into something customers rely on.
• A small, senior team. Few meetings, fast decisions, and code review from people who care how things work.
• Room to grow. Chance to be part of the early founding team. As Thinkbay grows, early engineers grow into tech leads and owners of whole products.

How to apply
Send your CV and a link to your GitHub or one project you are proud of. Tell us in a few lines about an agent or LLM system you built: what it did, what broke, and how you fixed it.
Our process: a short intro call, a technical round on Python and DSA, a practical round where you design or debug an agent, and a final conversation with the founders.
Location: Remote · Type: Full-time contract · Openings: 2