Role Overview:This role is for an AI Backend Engineer, also described as a Python Full Stack Developer, with strong Python and AI/ML fundamentals. The position focuses on hands-on experience in prompt engineering, integrating pretrained/foundation models into enterprise applications, and deploying AI solutions using CI/CD. The engineer will be versatile in constructing AI pipelines using tools like LangChain, LlamaIndex, and AutoRAG, with an emphasis on model optimization through Optuna and GridSearchCV.
Key Responsibilities:- Implement Python microservices and APIs (FastAPI/Flask/Django) with clean contracts, versioning, pagination, and rate limiting.
- Build REST/GraphQL endpoints and internal SDKs to enable AI enhanced features (prompt routing, retrieval, redaction) while maintaining strict separation from model ownership.
- Implement asynchronous and event driven processing (e.g., Celery/Kafka/queues) for high throughput pipelines and background jobs.
- Integrate services with enterprise AI platforms / LLM APIs (e.g., Azure OpenAI via enterprise gateway), handling prompt orchestration, tool invocation, moderation hooks, and response validation.
- Enforce runtime guardrails: input sanitization, PII redaction, toxicity/NSFW filters, output signing, and fallbacks.
- Implement semantic retrieval and vector search via approved platforms; do not train or host custom models.
Required Skills:- Strong Python and AI/ML fundamentals.
- Hands-on experience in prompt engineering.
- Experience integrating pretrained/foundation models into enterprise applications.
- Proficiency in deploying AI solutions using CI/CD.
- Versatility in constructing AI pipelines using LangChain, LlamaIndex, and AutoRAG.
- Experience with model optimization through Optuna and GridSearchCV.
- Familiarity with Python microservices and APIs (FastAPI/Flask/Django).
- Knowledge of REST/GraphQL endpoints and internal SDKs.
- Experience with asynchronous and event-driven processing (e.g., Celery/Kafka/queues).
- Ability to integrate with enterprise AI platforms / LLM APIs (e.g., Azure OpenAI).
- Understanding of prompt orchestration, tool invocation, moderation hooks, and response validation.
- Capability to implement runtime guardrails including input sanitization, PII redaction, toxicity/NSFW filters, output signing, and fallbacks.
- Experience with semantic retrieval and vector search.
Qualifications:- No specific years of experience mentioned.
Preferred Skills: 
