Llm Engineer

Summary

Join IOHK's Technical Intelligence team as an experienced AI/ML engineer to enhance our automated blockchain industry reconnaissance system. You will leverage large language models (LLMs) to optimize our data processing pipeline, build AI agents for in-depth report generation and technical architecture analysis, and improve NLP classification. This role requires expertise in LLMs, AI agents, NLP, and Python's data stack. You will collaborate closely with leadership to define requirements and contribute to the department's strategy. IOHK offers remote work, laptop reimbursement, a new starter package, learning and development opportunities, and competitive PTO.

Requirements

• BSc/MSc in Computer Science or a related field, preferably with a focus on NLP, AI, or ML
• 4+ years of work experience in engineering ML or AI solutions
• Experience with cloud platforms, preferably Databricks and AWS
• Experience with LLM toolstack, for example : Openai API, Auto-GPT, Langchain, Llamaindex, Chromadb, or other vector databases
• Fine-tune (augment) an LLM model with support of open source tooling such as Langchain, Llamaindex, Vectorized Databases, or Graph Databases
• Solid knowledge of LLM scaling systems, such as vllm, DeepSpeed, Ray
• Solid knowledge of AI agent systems such as Autogen or CrewAI
• Good foundation in Machine Learning (ML), Natural Language Processing (NLP), and Deep Learning (DL)
• Solid knowledge of Python’s data stack with a focus on NLP: Spark, Spark-NLP, Regex, Pandas, Polars, Tensorflow, Pytorch, SpaCy, Transformers, Huggingface
• Solid programming skills covering algorithms, data structures, design patterns, asynchronous programming, and relational databases
• Solid software development practices covering version control, testing, and continuous integration
• Knowledge of the Blockchain domain and passionate about gaining further breadth and depth of the domain

Responsibilities

• Increase the capabilities of our automated system by exploiting large language models (LLMs)
• Integrate LLMs into our recognition processing pipeline
• Understand the manual analytical work of intelligence engineers and propose optimizations by using LLMs and AI agents
• Drive customizations of LLMs with finetuning or RAG from several data sources, including knowledge graph databases and search API, and iteratively experimenting, optimizing, and evaluating LLM metrics
• Build AI agents that can create deep-dive and state-of-the-art reports on emerging technical trends in the blockchain space
• Build AI agents that can extract and evaluate technical architectures from projects’ source code repositories
• Build AI agents that can assess the strength of engineering (coding) practices from source code repositories
• Use LLMs with prompt engineering for NLP classification in a legacy data system
• Set up lightweight UI/UX for interactions with LLM or AI agent outputs
• Work closely with leadership to understand and define requirements, ensuring alignment with the department’s strategy and roadmap

Benefits

• Remote work
• Laptop reimbursement
• New starter package to buy hardware essentials (headphones, monitor, etc)
• Learning & Development opportunities
• Competitive PTO