About the role
AI summarisedPlay a pivotal role in designing and delivering reliable, robust AI applications, algorithms, and frameworks that elevate the quality and performance of product offerings in healthcare. You will collaborate with a dedicated team to revolutionize diagnostics through low-cost and high efficiency systems.
Life SciencesOnsite
Key Responsibilities
- Architect, build, and deploy LLM-powered agent systems (chatbots, copilots, agents) ensuring safety, speed, and cost efficiency.
- Own the entire product development lifecycle: build $\rightarrow$ prototype $\rightarrow$ evaluate $\rightarrow$ harden $\rightarrow$ monitor.
- Build and maintain retrieval-augmented generation (RAG) pipelines, including indexing, chunking, embeddings, reranking, and grounding.
- Apply advanced context engineering techniques such as prompt design, tool calling, memory management, and compression.
- Design and deploy production-grade chatbots with multi-turn conversation flows, escalation mechanisms, and integrated safety guardrails.
- Implement critical risk controls including safety filters, jailbreak resistance, PII redaction, and abuse detection.
- Optimize system performance regarding latency, efficiency, token/cost budgets, streaming, and caching.
- Establish rigorous evaluation frameworks using golden sets, RAG/grounding scores, toxicity checks, and A/B testing.
- Collaborate with cross-functional teams (software, hardware, data science) to ensure algorithm integration into production systems.
- Mentor junior AI engineers and advocate for best practices in LLM engineering, including reproducibility and transparency.
Requirements
- 5+ years of hands-on experience in production-level chatbots development.
- At least 1 year of dedicated experience building LLM-based agents.
- Expertise with major LLMs/APIs (OpenAI, LangChain, Anthropic, Hugging Face, etc.) and deep knowledge of prompt/context engineering.
- Deep experience with RAG pipelines, vector/hybrid search (e.g., FAISS, pgvector, Pinecone), rerankers, and grounding techniques.
- Proficiency in Python is mandatory for development tasks.
- Demonstrated ability to develop resilient chatbots with state management, tool integration, and fallback strategies.
- Hands-on experience managing AI agent safety controls (content moderation, policy enforcement, hallucination mitigation).
- Experience profiling AI agent system performance (batching/streaming, async/concurrency, cost/latency budgeting).
- Experience evaluating LLMs using frameworks like RAGAS or G-Eval, and familiarity with A/B testing.
- Experience managing the full lifecycle of LLMs in production, including versioning and CI/CD deployments.