AI Engineer – Small Language Models (SLM)

About the role

Avathon is seeking an experienced AI Engineer to design, fine‑tune, and deploy Small Language Models (SLMs) that power production‑grade AI features for supply‑chain applications. The role sits at the intersection of data science, product development, and MLOps, delivering real‑time intelligence to large enterprises.

Key responsibilities

• Build and fine‑tune SLMs/LLMs for use cases such as semantic search, forecasting, contract intelligence, and conversational insights, optimizing for performance, cost, and latency.
• Embed supply‑chain domain knowledge into models by collaborating with experts and product teams, and enable agentic workflows for planning and decision support.
• Integrate models into Avathon’s graph‑based micro‑services architecture using custom Lambdas and vector‑search techniques.
• Deploy, monitor, and operate models on cloud platforms (AWS, GCP, Azure) following strong MLOps practices.
• Continuously evaluate model quality to reduce hallucinations and improve reliability.

Required profile

• 8+ years of experience in forecasting, optimization, supply‑chain analytics, or AI‑driven product development.
• Proven experience with Small Language Models (e.g., Mistral, Phi, Nemotron, Llama variants) and parameter‑efficient fine‑tuning methods such as LoRA/QLoRA.
• Strong understanding of supply‑chain data workflows and familiarity with Knowledge Graphs and Neo4j.

Required skills

• Python
• PyTorch or TensorFlow
• GraphQL
• Neo4j
• FAISS (vector indexes)
• AWS, GCP, Azure
• MLOps practices
• LoRA / QLoRA fine‑tuning
• Retrieval‑Augmented Generation pipelines