Machine Learning Engineer

About the Role

TechNova's AI Engineering practice takes client requests like "help our support team answer questions faster" or "let underwriters query policy documents in plain English" and turns them into production systems — not demos. As a Machine Learning Engineer, you'll design and ship these systems end-to-end, from data pipeline through evaluation to deployment.

This role sits at the intersection of applied ML and solid software engineering: you'll be expected to reason about model selection and prompt design just as rigorously as you reason about latency, cost, and failure modes in production.

Responsibilities

• Design and build retrieval-augmented generation (RAG) pipelines using vector databases and embedding models
• Integrate large language models (OpenAI, Claude, and open-weight models) into client applications via API and self-hosted deployments
• Build evaluation harnesses to measure model output quality, latency, and cost before and after production changes
• Design data ingestion and chunking pipelines for unstructured client data (documents, support tickets, transcripts)
• Collaborate with full-stack engineers to integrate AI features into existing product surfaces
• Monitor production AI systems for drift, hallucination rates, and cost, and iterate accordingly
• Advise clients on realistic AI use cases versus overhyped ones, grounded in what's actually production-ready

What We're Looking For

• 5+ years of software engineering experience, with 2+ years focused on applied ML or LLM-based systems
• Strong Python skills, including experience with frameworks like LangChain, LlamaIndex, or comparable orchestration tooling
• Hands-on experience with vector databases (e.g., Pinecone, Weaviate, pgvector) and embedding-based retrieval
• Practical understanding of LLM behavior: prompt engineering, context window management, function/tool calling, and failure modes
• Experience deploying ML-backed features to production on AWS or similar cloud infrastructure, with proper monitoring and cost controls
• Ability to communicate technical trade-offs to non-technical client stakeholders across a distributed, global team

Nice to Have

• Experience fine-tuning or evaluating open-weight models
• Background in a regulated industry where explainability and auditability of AI outputs matter
• Contributions to ML/LLM open-source tooling or published technical writing