Senior GenAI Engineer

Senior GenAI Engineer Applied LLMs

Job Type: Full-time

Experience: 4 - 7 years

About the Role

We are looking for a hands-on Senior GenAI Engineer to help us design, build, and scale production-grade Generative AI and Agentic AI systems. We welcome candidates from diverse technical backgrounds - whether

you are a software engineer who grew into AI, or a data scientist who transitioned away from classical ML.

What truly matters is your ability to engineer reliable, trustworthy LLM-powered products from start to finish, including robust agent workflows, memory, retrieval, evaluation, and cloud deployment.

What You'll Do

• Design and deploy real-world GenAI applications, including RAG systems, copilots, agentic workflows, and multi-agent solutions.
• Build applications using both commercial APIs (OpenAI, Anthropic Claude, Google Gemini) and open weight models (Llama, Mistral, DeepSeek).
• Build and orchestrate AI agents using modern agentic AI frameworks, including tool/function calling, structured outputs, planning, routing, and human-in-the-loop patterns.
• Design memory and state-management patterns for agentic applications, including session/short-term memory, persistent/long-term memory, context handling, and workflow checkpointing.
• Architect robust retrieval pipelines using vector databases such as Pinecone, Qdrant, Weaviate, or pgvector, including chunking, embeddings, semantic/hybrid search, and re-ranking.
• Set up rigorous evaluation frameworks such as LLM-as-a-judge or RAGAS to monitor accuracy, hallucinations, latency, reliability, and costs.
• Deploy and operate AI services on at least one major cloud platform - AWS, Azure, or GCP - using Docker and CI/CD practices.
• Write clean, modular, well-documented code while collaborating with product teams and mentoring junior engineers.

What We're Looking For

• 4 - 7 years of overall engineering, data science, ML, or AI experience, with at least 1.5+ years deeply focused on building GenAI applications for production.
• Strong proficiency in Python and modern AI application development, with hands-on experience building APIs/services and production-ready solutions.
• Real-world experience with RAG pipelines, chunking strategies, embeddings, vector databases, and hybrid or semantic search.
• A solid understanding of LLM behavior, including context-window management, prompt leakage/drift, hallucination mitigation, and reliable structured outputs.

Must-Have Skills

1. Tools - Foundational: Hands-on comfort with the core tools used to build and ship GenAI applications, including Python, Git/version control, REST APIs, Docker, and CI/CD workflows.

2. Skills - Foundational: Strong software/AI engineering fundamentals: clean and modular coding, debugging, testing, API integration, prompt engineering, structured outputs, and function/tool calling.

3. Memory Management: Hands-on experience implementing memory in agentic or conversational AI systems - such as session/short-term memory, long-term/persistent memory, conversation state,

checkpointing, context persistence, or memory retrieval/update strategies.

4. Cloud: Hands-on deployment experience on at least ONE of Azure, AWS, or GCP. AWS is good to have but is NOT mandatory.

5. Agentic AI Frameworks - Mandatory: Hands-on experience with at least ONE of the following frameworks is required:

• LangGraph
• LangChain
• CrewAI
• AutoGen (Microsoft)
• OpenAI Agents SDK
• Google ADK (Agent Development Kit)
• AWS Strands Agents
• PydanticAI
• Agno (formerly Phidata)
• LlamaIndex
• Semantic Kernel (Microsoft)
• Haystack

Note: Listing a framework as a skill is not enough; candidates should be able to explain where and how they

used it in an agentic workflow or production-oriented application.

Bonus Points If You Have

• Familiarity with fine-tuning techniques such as LoRA/QLoRA or inference optimization.
• Hands-on experience with Model Context Protocol (MCP), agent-to-agent (A2A) patterns, guardrails, or agent observability/evaluation.
• Experience designing advanced multi-agent patterns such as planner-executor, supervisor/router, critic reflector, or human-in-the-loop workflows