We are seeking a talented and motivated Software Student to join our Algorithm Infrastructure team, developing internal applications that empower advanced data management and algorithm research.
This role offers the opportunity to work across the full stack using modern technologies such as Python and React, and to go a step further by building and operating agentic AI infrastructure: MCP servers, LLM-powered workflows, and internal AI tooling used daily by engineers across the organization.
*
What's in it for you?
Join a collaborative and technically engaged environment where you will:
– Work hands-on with full-stack technologies (Python, React) and modern development practices.
– Build impactful applications used daily by algorithm engineers for research and data workflows.
– Design and ship MCP (Model Context Protocol) servers that extend AI coding assistants with custom capabilities — from semantic code search to sprint monitoring.
– Develop and maintain an internal Claude Code marketplace, contributing plugins and agents used by engineers across the organization.
– Build agentic pipelines — automated PR reviewers, LLM-driven data processors, sprint analyzers — that replace manual toil with intelligent automation.
– Work with vector search, embeddings, and multi-model LLM routing (Anthropic Claude, OpenAI, VoyageAI) in production systems.
– Be part of a team that encourages independence, curiosity, and continuous learning.
*
What you'll be doing:
– Design and develop full-stack applications (Python/FastAPI backend, React/TypeScript frontend) to support data and algorithm workflows.
– Build and maintain MCP servers that surface internal data, tools, and services to AI coding assistants such as Claude Code — enabling engineers to work faster through context-aware AI tooling.
– Contribute to the team's internal AI plugin marketplace: develop, version, and publish Claude Code skills and agents (PR reviewer, product owner, sprint analyzer, report generators) consumed by engineering teams.
– Implement agentic workflows — multi-step AI pipelines that autonomously gather context, reason over data, and produce structured outputs — using frameworks built in-house and open-source tooling.
– Work with vector stores, semantic embeddings, and code intelligence pipelines (e.g., Tree-sitter AST parsing, 32-language semantic search) to build tools that understand large codebases.
– Build systems that help algorithm engineers manage, analyze, and interact with large-scale datasets, including data upload/indexing services and self-service research platforms.
– Work closely with algorithm teams to understand their needs and translate them into effective tools — including AI-augmented ones.
– Contribute hands-on to both architecture and implementation across microservices, container orchestration (Docker, Kubernetes), and CI/CD pipelines.
*
What you should have:
– Current Computer Science student with at least 1.5 years remaining until graduation.
– Experience in Python and/or React (full-stack orientation is a strong advantage).
– Curiosity about — or hands-on experience with — LLMs, agentic AI patterns, and AI-augmented development workflows (e.g., Claude Code, GitHub Copilot, Cursor).
– Familiarity with development tools and environments such as Git, Docker, and Kubernetes.
– Ability to work at least 3 days a week.
– Strong problem-solving skills, with the ability to work independently.
– Curious, self-driven, and eager to learn new technologies.
– Strong English communication skills, both written and verbal.
Nice to have (you'll pick these up on the job):
– Exposure to the Model Context Protocol (MCP) or building tool-use integrations for LLMs.
– Experience with vector databases, embeddings, or semantic search (DuckDB, FAISS, Chroma, etc.).
– Understanding of prompt engineering, multi-step agentic pipelines, or LLM orchestration frameworks.
– Experience with FastAPI, async Python, or WebSocket-based real-time systems.
*
Who we are?
We are a dynamic and innovative team dedicated to building infrastructure that accelerates algorithm development and data-driven insights. Our work bridges the gap between software engineering and advanced research — and increasingly, between human engineers and AI agents.
We don't just use AI tools, we build the infrastructure layer that makes AI coding assistants useful at scale inside the organization: custom MCP servers, internal plugin marketplaces, agentic review pipelines, and LLM-powered data workflows.
If you want to shape how a real engineering organization integrates agentic AI into its daily work — not as a user, but as a builder — this is the team for you.


