Artificial Intelligence Engineer

THE OPPORTUNITY AllysAI is not a typical AI company. We're an AI Lab-as-a-Service that helps enterprises escape "pilot hell" and ship production AI in 60-90 days—not 12+ months. Our clients include Al Futtaim, Merz Pharmaceuticals, Abu Dhabi Government entities, and Sanofi. We don't do PowerPoint AI. We build systems that run in production, handle real workloads, and deliver measurable ROI. We're looking for an AI Engineer who doesn't just write code—you architect intelligent systems. You're the person who reads AI papers on weekends, has opinions about which vector database to use and why, and can explain transformer attention mechanisms to a CEO in plain English. You ship fast, iterate faster, and treat every project like it's your startup. This isn't a job. It's a launchpad. We're building the team that will define how enterprises adopt AI in the Middle East and beyond. The right person will grow into a CTO role as we scale. WHAT YOU'LL DO Build Production AI Systems (not experiments) Design and deploy RAG architectures, multi-agent systems, and LLM-powered applications for enterprise clients Build process automation pipelines that replace manual workflows with intelligent agents Implement fine-tuning pipelines when off-the-shelf models don't cut it Ship solutions in 4-8 week sprints that meet enterprise security, scalability, and compliance standards Own the Technical Vision Evaluate and integrate emerging AI tools, frameworks, and models weekly Make architecture decisions that balance speed, cost, and performance Build reusable components and internal tools that accelerate future projects Document systems so they can be maintained, extended, and handed off to clients Bridge Tech and Business Join client calls to understand requirements and translate them into technical specs Explain tradeoffs to non-technical stakeholders without dumbing it down Identify opportunities where AI can solve problems clients didn't know they had Contribute to proposals, scoping, and technical due diligence for new engagements Move at Startup Speed Prototype solutions in days using vibe coding tools (Claude, Codex, Gemini, Lovable, Cursor) Deploy to production environments confidently and frequently Debug, iterate, and improve based on real-world feedback Work autonomously while keeping the team aligned on progress and blockers WHO YOU ARE AI-Obsessed You don't just use AI—you study it. You follow AI Twitter/X, read papers, experiment with new models the week they drop, and have strong opinions about prompt engineering, RAG chunking strategies, and agent architectures. AI isn't your job; it's your obsession. Technically Exceptional You're a Python beast. You've built production systems, not just Jupyter notebooks. You understand software engineering fundamentals: clean code, version control, testing, CI/CD, monitoring. You can debug a LangGraph agent at 2am and enjoy it. Business-Minded Builder You understand that code is a means to an end. You think about user problems, ROI, and adoption. You can scope a project, estimate effort, and make tradeoffs that balance technical excellence with business reality. Future CTO Material You want to lead. You think about system design, team scaling, technical strategy, and how to build something lasting. You're not looking for a comfortable job—you're looking for a rocket ship. REQUIRED SKILLS & EXPERIENCE Experience 3+ years building production software (not just ML experiments) 1+ years working specifically with LLMs, RAG, or generative AI in production Track record of shipping: show us what you've built Core Technical Skills Python: Expert-level. FastAPI, Pydantic, async programming, clean architecture LLM Orchestration: LangChain, LangGraph, LangSmith, LlamaIndex—you've built with at least two RAG Systems: Vector databases (Pinecone, Weaviate, Qdrant, Chroma), embedding strategies, retrieval optimization, chunking approaches Agent Frameworks: CrewAI, AutoGen, custom agent architectures—you understand when to use agents and when they're overkill Automation Platforms: n8n, Make, Relevance AI, or similar—you can build workflows that connect AI to business systems APIs & Integrations: REST, webhooks, OAuth—you can connect anything to anything Cloud Platforms: AWS, GCP, or Azure—you can deploy, scale, and monitor AI/ML Depth Prompt engineering: You write prompts that work, not prompts that sort of work Fine-tuning: You've fine-tuned models (LoRA, QLoRA, or full fine-tune) and know when it's worth it Evaluation: You can measure AI system performance beyond "it looks good" Cost optimization: You understand token economics and can architect for efficiency Vibe Coding Proficiency You actively use AI coding assistants: Claude, Codex, Cursor, Gemini, Windsurf, Lovable You can scaffold entire applications in hours, not days You know how to prompt coding assistants for maximum output quality You understand the limits—when to trust AI-generated code and when to rewrite Bonus Points Experience with multi-modal AI (vision, audio, video) Knowledge of MLOps: model versioning, experiment tracking, deployment pipelines Familiarity with enterprise requirements: SOC2, GDPR, data residency Open source contributions or public projects in AI space Experience in consulting, agencies, or client-facing technical roles TECH STACK YOU'LL WORK WITH Languages & Frameworks Python, FastAPI, Pydantic, TypeScript (for frontend integrations) LLM & AI OpenAI, Anthropic Claude, Google Gemini, Mistral, Llama, Cohere LangChain, LangGraph, LangSmith, LlamaIndex, CrewAI, AutoGen Vector & Data Pinecone, Weaviate, Qdrant, Chroma, PostgreSQL + pgvector, Supabase Automation & Integration n8n, Make, Relevance AI, Zapier, custom webhook architectures Infrastructure AWS (Lambda, ECS, Bedrock), GCP (Cloud Run, Vertex AI), Azure, Vercel, Railway Docker, GitHub Actions, Terraform (nice to have) Vibe Coding Claude, Cursor, GitHub Copilot, Windsurf, Lovable, v0, Bolt HOW WE WORK Remote-first: Work from anywhere with 4+ hours overlap with UAE timezone (GMT+4) Async communication: We document decisions, write clear specs, and respect deep work time High ownership: You own projects end-to-end, from scoping to deployment to client handoff Fast feedback loops: Weekly demos, rapid iteration, direct client interaction Learning culture: We share what we learn, experiment with new tools, and level up together SUCCESS METRICS First 30 Days Complete onboarding and understand AllysAI's existing systems and client projects Ship your first feature or component to a live client project Demonstrate proficiency with our core stack (LangGraph, n8n, vector DBs) Establish your development workflow and communication rhythm First 90 Days Lead a client project module from technical design to deployment Build at least one reusable internal tool or component Contribute to technical scoping and estimation for new proposals Demonstrate ability to work directly with clients on technical discussions First 6 Months Own end-to-end delivery of a major client engagement Introduce at least one new tool, framework, or approach that improves our delivery Begin mentoring junior team members or contractors Contribute to AllysAI's technical strategy and roadmap First Year Recognized as technical leader across multiple client engagements Built systems serving enterprise clients with measurable business impact Clear trajectory toward Lead Engineer or CTO track Shaped AllysAI's technical culture and standards COMPENSATION Base Salary: Competitive, based on experience and location Performance Bonus: Tied to project delivery and client outcomes Equity: Meaningful stake in AllysAI's growth Learning Budget: Resources for courses, conferences, and tools Equipment: What you need to do your best work WHY ALLYSAI Impact from Day One No six-month onboarding. No bureaucracy. You'll ship to production in your first week and see your work used by real enterprises. Work on Hard Problems Enterprise AI implementation is genuinely difficult. You'll solve problems that don't have Stack Overflow answers—and build expertise that's rare in the market. CTO Track We're explicit about this: we're looking for someone who wants to lead. As AllysAI grows, so does your role. The path from AI Engineer to CTO is real and achievable here. Front Row to the AI Revolution You'll work across industries (pharma, real estate, government, finance), use cases (sales, marketing, HR, operations), and technologies (RAG, agents, automation, fine-tuning). No two projects are the same. Founder Access You'll work directly with the CEO. No layers. No politics. Direct feedback, direct impact, direct growth. HOW TO APPLY Send an email to hiring@allysai.com with subject line: "AI Engineer - [Your Name]" Include: Your resume/CV (PDF) Portfolio or GitHub showing AI projects you've built (required) Short Loom video (3-5 min): Walk us through an AI system you built—architecture decisions, challenges, what you'd do differently Answer these questions (in the email body): What's the most interesting AI paper, tool, or development you've encountered in the last month? Why? Describe your ideal tech stack for building a RAG-based customer support agent. Justify your choices. What's your controversial opinion about the current state of AI engineering? No portfolio = no review. We need to see what you've built. INTERVIEW PROCESS Application Review (3-5 days) Technical Screen (30 min): Quick chat about your background and a few technical questions Technical Deep Dive (60-90 min): Architecture discussion, code review of your work, live problem-solving Paid Trial Project (1-2 days): Real task, real compensation—see how we work together Final Conversation (30 min): Culture fit, questions, offer discussion Total timeline: 2-3 weeks from application to offer. AllysAI is building the future of enterprise AI adoption. If you want to be part of defining how the world's largest companies implement AI, apply now.