Senior Backend Developer

We are looking for a Senior Backend Developer with deep expertise in Generative AI and LLM Systems to

design, build, and scale the intelligent backbone of our products. You will bridge solid software

engineering fundamentals with cutting-edge AI capabilities — owning everything from RESTful API design

and system architecture through to LLM orchestration, multi-agent workflows, and RAG pipelines. You will

work closely with product, data science, and front-end teams, bringing both technical depth and a

pragmatic, delivery-focused mindset to every project.

• Architect and develop scalable, production-grade backend services and APIs that power AI-driven features.
• Design and implement LLM orchestration pipelines using frameworks such as LangChain, LlamaIndex, or custom-built solutions.
• Build, evaluate, and maintain multi-agent systems capable of handling complex, multi-step task orchestration.
• Engineer, test, and continuously optimise prompts for accuracy, reliability, consistency, and cost efficiency.
• Implement and maintain Retrieval-Augmented Generation (RAG) pipelines, including document ingestion, chunking, embedding, and retrieval strategies.
• Integrate AI models with internal and third-party APIs, vector databases, relational databases, and external tools.
• Fine-tune, evaluate, and monitor LLMs using domain-specific datasets; establish robust evaluation frameworks to track model quality over time.
• Implement guardrails, safety checks, and alignment strategies to reduce hallucinations and ensure responsible AI outputs.
• Proactively identify performance bottlenecks, technical debt, and opportunities to improve system reliability and developer experience.
• 5–7 years of professional backend / full-stack development experience, with at least 2 years focused on AI or ML systems.
• Strong Python proficiency — clean, testable, production-ready code.
• Deep, hands-on expertise in prompt engineering: chain-of-thought, few-shot, system prompts, structured outputs, and tool use.
• Proven experience building with LLMs in production — OpenAI (GPT-4o / o-series), Anthropic Claude, Mistral, LLaMA, or similar.
• Solid understanding of agentic patterns: ReAct, plan-and-execute, tool-calling agents, and agent memory strategies.
• Hands-on experience with LLM orchestration frameworks (LangChain, LlamaIndex, AutoGen, CrewAI, or equivalent).
• Experience designing and operating RAG pipelines end-to-end, including chunking strategies, embedding models, and re-ranking.
• Familiarity with vector databases and embeddings (Pinecone, Weaviate, Qdrant, FAISS, pgvector, or similar).
• Strong API design skills — RESTful services, versioning, authentication, rate limiting, and error handling.
• Experience with relational and NoSQL databases (PostgreSQL, MongoDB, Redis) and their use alongside AI workloads.
• Understanding of AI evaluation metrics (RAGAS, LLM-as-judge, latency, cost, groundedness) and
• model performance tuning.
• Comfortable working in cloud environments (AWS / GCP / Azure) and containerised deployments (Docker, Kubernetes).

Preferred Qualifications

• Experience fine-tuning LLMs (LoRA/QLoRA, SFT, RLHF) or training custom models on domain-specific datasets.
• Exposure to MLOps tooling — experiment tracking (MLflow, W&B), model registries, and automated evaluation pipelines.
• Experience building conversational AI products: chatbots, copilots, voice assistants, or similar.
• Knowledge of AI safety, guardrails, and governance frameworks (e.g., content moderation, PII detection, output filtering).
• Familiarity with streaming responses, WebSockets, or Server-Sent Events for real-time AI interfaces.
• Contributions to open-source AI projects or a portfolio demonstrating AI system design.
• Exposure to graph databases or knowledge graphs as a complement to vector retrieval.

TECHNOLOGY STACK

Core technologies you will work with day-to-day:

• Python • Docker & Kubernetes

• OpenAI / Anthropic / OSS LLMs • AWS / GCP / Azure

• LangChain / LlamaIndex • GitHub Actions (CI/CD)

• Pinecone / pgvector / FAISS • Prompt engineering & eval frameworks

• PostgreSQL & Redis • MLflow / Weights & Biases