Senior AI Developer
Responsibilities:
- Leading the design and development of AI features (chatbots, agents, RAG pipelines, copilots) from discovery to production;
- Building high-quality React frontends and Python backends that integrate with internal systems;
- Implementing robust MLOps/LLMOps on Azure (CI/CD, observability, evaluation, safety/guardrails, cost controls);
- Architect reusable platform components: prompt/orchestration layers, vector search, data pipelines, feature stores, model registries;
- Translating business problems into technical designs; running lean experiments and iterating quickly with stakeholders;
- Establishing engineering standards (coding, testing, security, documentation) using GitHub and DevOps practices;
- Ensuring data privacy, governance, and compliance in all AI solutions;
- Measuring outcomes (quality, latency, adoption, ROI) and iteratively improving.
Requirements:
- At least 5 years of software engineering, with recent focus on AI/ML or LLM-based products;
- Expert in Python (APIs, data processing, async, testing) and React (modern hooks, state management, performance);
- Proven delivery of generative-AI applications (LLM orchestration, prompt design, evaluation, retrieval/RAG, tool/agent patterns);
- Hands-on with Microsoft Azure (App Services/AKS/Functions, Azure OpenAI or equivalents, Azure Storage, Key Vault, Monitor);
- Strong CI/CD using GitHub (Actions), Infrastructure-as-Code, and DevOps practices;
- Solid understanding of data and systems design (APIs, events/queues, microservices, caching, observability);
- Pragmatic product mindset: ability to scope MVPs, validate value fast, and communicate clearly with non-technical partners;
- Security-first approach (authN/Z, secrets, data protection, safe-use policies for AI).
Nice to have:
- Experience with LLM frameworks (LangChain, Semantic Kernel), vector databases (Chroma, Pinecone, FAISS), and evaluation tooling;
- Classic ML/DS experience (feature engineering, model training/serving, monitoring);
- Prompt/guardrail techniques, content moderation, and bias/robustness testing;
- Event-driven architectures (Kafka/Service Bus), GraphQL, or WebSockets for real-time UX;
- Analytics/telemetry to measure product impact.