We are looking for a mid-level Data Scientist / Python Developer (2–5 years experience) to help us bring a RAG–agentic platform from “works in dev” to “works in the real world”.
It’s a hands-on, experimental position at the intersection of data science, LLM evaluation, and backend Python development.
You’ll simulate users, design experiments, analyze results, and contribute code to make the platform smarter and more reliable.
You’ll be the person who:
Puts the LLM & RAG platform under real-life scenarios,
Evaluates how it behaves with LLMs and different data sources, and
Integrates open-source components and tools into the platform, and
Extends parts of the system with new ideas and features that can later be productized.
What You’ll Do
Design and run realistic usage scenarios to evaluate the RAG / agentic platform (simulate real users, edge cases, tricky queries, etc.).
Measure and analyze platform performance (quality of responses, latency, failure modes, regressions) using logs, metrics, and your own experiments.
Prototype and extend system functionality in Python (new tools/agents, prompts, pipelines, evaluation scripts, small services) that can be integrated into the platform.
Collaborate with the core engineering team to report findings, suggest improvements, and iterate on system behavior.
Work with LLMs and RAG pipelines: prompt and retrieval tuning, evaluating different models/configurations, and testing integrations with external data sources.
Build and maintain test/evaluation suites that can be reused to validate new platform versions.
Write clean, well-documented, and testable Python code for experiments, internal tools, and platform extensions.
What We’re Looking For
Strong Python skills – you are comfortable writing idiomatic, modular code, working with virtual environments, using type hints, handling errors properly, and writing basic tests (pytest or similar).
Hands-on experience using Python for data and backend tasks: data processing (pandas/NumPy or similar), scripting and automation, calling external APIs, and working with files / structured data (JSON, CSV, Parquet, etc.).
2 - 5 years of professional experience as a Data Scientist, ML Engineer, or Python Developer working in a production or near-production environment.
Experience with data analysis / experimentation: running experiments, comparing variants, interpreting results, and turning findings into concrete code or configuration changes.
Familiarity with LLMs or modern ML tooling (e.g. OpenAI, Hugging Face, vLLM, or similar APIs/frameworks), ideally used from Python.
Understanding of web services and APIs (REST, JSON, authentication) and how to consume or expose them in Python (e.g. FastAPI/Flask or similar).
Comfortable working in a remote, async-friendly environment, sharing progress in writing, and communicating clearly with engineers and stakeholders.
Nice to Have (Bonus, Not Mandatory):
Hands-on work with RAG systems, agentic workflows, or vector databases.
Experience with evaluation of LLM-based systems (e.g. prompt benchmarking, automatic or human-in-the-loop evaluation).
Familiarity with common Python data/ML stack (pandas, NumPy, scikit-learn, PyTorch, etc.).
Experience with backend development (FastAPI / Flask / Django, background tasks, integration scripts).
Contributions, personal projects, or a GitHub profile showing interest in AI / LLMs / data-heavy systems.
Application Closing Date
31st December, 2025.
How to Apply
Interested and qualified candidates should send their CV and GitHub / Portfolio to:careers.ogesoft@gmail.com using the Job Title as the subject of the email.