About the Role
As an AI Engineer at Upvista, you will implement context-aware reasoning engines, fine-tune retrieval pipelines, design evaluations, and build sandboxed execution containers. You bridge the gap between LLM model research and robust production systems that handle real work without failing.
Requirements
- Active GitHub contributions showing real AI experiments, agent systems, or LLM-driven applications.
- Experience with Python, Node.js, and raw LLM API integrations (OpenAI, Anthropic, Gemini).
- Firm grasp of vector databases (Pinecone, pgvector), retrieval-augmented generation (RAG), and embedding structures.
- Experience building sandboxed environments (Docker, firecracker) for secure agent code execution.
- Strong understanding of prompt engineering, function calling, structured outputs, and evaluation metrics.
Responsibilities
- Build and deploy autonomous agent systems that handle coding, editing, search, or workflow execution.
- Optimize context retrieval pipelines to maximize accuracy and minimize model latency/cost.
- Design sandbox computation engines to safely execute code written by agents.
- Conduct evaluation runs to track agent performance, regression, and model consistency.
Apply for this role
Please fill out the details below. We value clear, direct communication, high autonomy, and a proven track record of shipping production software.
YOUR BACKLOG, OUR PROBLEM
What’s sitting on your engineering backlog?
Give us one task. We’ll take it from scope to production while your team keeps building the core product.