AI ENGINEERING | AGENTIC AI & LLM SYSTEMS
We are working with technology organisations building at the frontier of large language models and agentic AI — spanning production-grade AI Agent products for enterprise platforms, and foundational research into large-model training, inference acceleration, and multi-agent systems. This posting covers a family of AI Engineering roles across both applied product engineering and research-adjacent infrastructure work.
We are looking for engineers who want to build agentic systems from the ground up rather than simply integrate an existing API. This is hands-on technical work with real scope — you will help turn evolving product and research requirements into working agent systems, evaluation pipelines, and, depending on the track, core model/training infrastructure that other teams build on.
- Design, build, and iterate on AI Agent capabilities — orchestration, tool/function calling, RAG, multi-step workflow automation — for a core enterprise product, working closely with Product to turn requirements into production features.
- On the research/infrastructure track: build core large-language-model competitiveness — long-context reasoning, agentic capabilities, training/inference acceleration (quantization, sparsity, distributed parallelism, kernel-level optimization) — and architect scalable agent frameworks spanning planning, memory, and execution.
- Build evaluation pipelines and instrumentation to continuously measure and improve agent accuracy, latency, reliability, and cost in production.
- Partner closely with Product, Engineering, Infra, and (on the research track) contribute to technical publications, open-source projects, or patents.
- Use data and production feedback to identify failure modes, improve reliability, and continuously refine how the agent systems perform as usage scales.
Your current title could be AI Engineer, Senior LLM Researcher, Machine Learning Engineer, Applied Scientist, or something else entirely. More important is what you have actually built, such as:
- Building or extending LLM-powered or agentic systems — through coursework, research, internship, or production work
- Working hands-on with agent frameworks (LangChain, LlamaIndex, AutoGen), RAG pipelines, or tool/function-calling architectures
- Contributing to large-model training, fine-tuning, or inference-acceleration work (distributed parallelism, quantization, sparsity)
- Shipping AI-powered features end-to-end, from prototype to production
- Publishing, open-sourcing, or patenting work related to LLMs, agents, or large-model infrastructure
You should be:
- Educated to Master's degree level or above in Computer Science, Artificial Intelligence, Data Science, Electrical/Computer Engineering, or a closely related computational field — this is a firm requirement for this posting.
- Comfortable with Python as a primary language, with familiarity in at least one of Java/Go/TypeScript/C++.
- Practically fluent in LLM concepts — prompting, embeddings, RAG, agents, tool/function calling — with at least foundational ML/DL knowledge.
- A fast, self-driven learner, comfortable picking up new frameworks and vector databases independently and operating with a bias for action in ambiguous problem spaces.
- PhD and a record of publications or open-source leadership are a plus for the research/infrastructure track, but not required.
This is an opportunity to work on agentic AI systems at a formative stage — rather than inheriting a fixed technical playbook, you will have real influence over how the agent architecture is designed, how model and infrastructure decisions are made, and how the AI engineering function scales as these products and research programmes grow.
If you are interested in the above opportunity, please forward your resume in discretion to [email protected] | Licence No. 18S9318 | R1875384