Software/Data Engineer (Python) - SPVL

Summary

Build and maintain AI agents for cybersecurity, tuning prompts, optimizing token costs, and deploying ML models in Python with cloud data pipelines.

We are looking for Software/Data Engineer to work on AI project within Cybersecurity domain:

Responsible for managing database and server rack services supporting agentic solutions, run analysis, grant token access, re‑train/refine the model and prompt‑engineering.
Work location: Central (hybrid work arrangement).
Entry level candidate with Cybersecurity related experience or internship are welcome to apply.

Write and tune system prompts for AI agents following standards from the Senior AI Software Engineer.
Reduce token cost using caching, batching, and routing to the right models.
Set up production monitoring: tracing, cost and latency tracking, quality dashboards, and release checks.
Improve model speed and deployment pipelines for production.
Build and maintain tests for AI output quality – accuracy checks, regression tests, and reference question/answer sets.
Implement fine‑tuning workflows (e.g. LoRA/QLoRA) under senior guidance.
Implement ML models and AI features for team use‑cases.
Research and prototype new AI/LLM tools for CDOI security work.
Prepare, version, and refresh datasets for RAG and fine‑tuning – complete, accurate, and up to date.
Maintain data systems for analytics, reporting, and AI across environments.
Build pipelines: ingest data, parse, enrich, split text, embed, and load into vector stores and warehouses.

Minimum 4 years of experience in software engineering, data engineering, MLOps, or applied AI in production.
Experience in building Python services/APIs and production data pipelines.
Strong in Python and SQL on cloud analytics warehouses.
Experience with LangChain, LlamaIndex, LangSmith, MLflow, or OpenTelemetry would be an added advantage.
Background or knowledge in Cybersecurity would be an added advantage.