Tech Stack
PythonFastAPIPyTorch/TransformersvLLMOllamaLangChain/LlamaIndexVector search (Qdrant/pgvector/Elasticsearch)PostgreSQLDockerKubernetesGitLab CI/CD
به این موقعیت علاقهمندی؟
رزومهات را برایمان بفرست تا بررسی کنیم.
About Wizdoma
Wizdoma builds Enterprise AI solutions, AI Assistants, RAG platforms, AI Agents, Document Intelligence, Workflow Automation for clients with demanding, often regulated requirements.
Responsibilities
- Own AI engineering as Wizdoma's first dedicated AI hire, set the roadmap, build the team as scope grows.
- Build production LLM services: document extraction/OCR, retrieval-based classification, and LLM-as-judge systems grounded in domain rules with auditable output as real APIs (async, idempotent, retries), not notebooks.
- Take services from prototype to production-hardened.
- Self-host and adapt open-weight models where commercial API access isn't reliable.
Requirements
- 3+ years shipping LLM systems to production.
- Strong Python; experience building async/idempotent APIs (FastAPI or similar).
- Hands-on with LLMs, RAG/semantic search, prompt engineering, structured/constrained output.
- Experience self-hosting models (vLLM, Ollama, or similar) — this role can't rely on OpenAI/Azure access.
- English proficiency.
- Docker, Kubernetes, CI/CD.
Nice to Have
- Fine-tuning open-weight models (LoRA/PEFT).
- RegTech/GovTech or compliance-domain experience.
- Speech-to-text/audio transcription.