Job Title: Senior Prompt Engineer (Strong Python Coding)
Location: Hyderabad – Onsite (5 Days)
Shift Timings: 2-11PM
Experience: 4+ Years (2+ in AI/LLMs)
Role Summary:
We are hiring a highly technical Prompt Engineer with strong Python backend expertise. This is not a content-writing role. This role requires deep understanding of LLM behavior, prompt optimization, guardrails, structured outputs, evaluation pipelines, and production-grade AI system design. If you cannot write scalable Python code, this is not the role.
Core Responsibilities:
Advanced Prompt Engineering:
- Design multi-step, system-level prompts.
- Build structured output prompts with strict JSON schema enforcement.
- Implement few-shot, chain-of-thought, self-consistency, tool-calling patterns.
- Reduce hallucinations using grounding strategies.
- Optimize token usage and latency.
Strong Python Development (Mandatory):
- Build AI microservices using FastAPI / Flask.
- Implement streaming responses and async pipelines.
- Write modular, production-grade Python code.
- Handle retries, fallbacks, rate limiting, caching.
- Integrate LLM APIs (OpenAI, Claude, Gemini, Bedrock).
Must be comfortable with:
- OOP in Python
- Async programming
- Error handling & logging
- Writing clean, testable code
Guardrails & Validation:
- Implement response validation layers.
- Build automated moderation and safety filters.
- Enforce schema validation before response delivery.
- Create deterministic output pipelines.
Evaluation Frameworks:
- Build automated evaluation pipelines.
- Define and measure:
- Accuracy
- Faithfulness
- Relevance
- Latency
- Cost per request
- Run A/B prompt testing.
RAG & Tool Integration:
- Implement Retrieval-Augmented Generation pipelines.
- Work with vector databases (FAISS, Pinecone, Weaviate).
- Build tool/function calling integrations.
- Design multi-agent workflows (LangChain / LangGraph).
Required Technical Skills:
- Expert-level Python
- LLM API integration
- JSON schema design
- Prompt debugging & optimization
- FastAPI / REST APIs
- Git & version control
- Cloud exposure (AWS / Azure / GCP)
Must-Have Experience:
- Building LLM systems in production
- Structured output enforcement
- Handling hallucination mitigation
- Working with evaluation frameworks
- Performance optimization
What We Do NOT Want:
- Candidates who only “use ChatGPT”
- Pure frontend developers
- Data scientists with no backend coding
- Prompt writers without engineering background
Preferred:
- Experience with healthcare/insurance AI systems
- Multi-agent orchestration
- MCP protocol exposure
- Voice/real-time AI systems