Location: Gurgaon
Reports To: CPTO
Experience Required: 6–8 years (with 3+ years in advanced AI/LLM/RAG systems)
About Group Bayport
Group Bayport, headquartered in Atlanta (USA), is a rapidly growing global e-Commerce B2B2C, B2B and B2C organization that has re-defined the business of delivering high-quality custom products through a unique blend of cutting-edge digital technologies, robust manufacturing capabilities and global supply chain. An industry leader in print technology, Group Bayport operates in the United States, Canada, Australia, New Zealand, UK, and India through its family of brands – BannerBuzz, Covers & All, Vivyx Printing, Circle One, Giant Media, Neon Earth, Optamark and Northcape.
We acquired Chicago based Northcape, a custom cushion and furniture company in August 2024, with a vision to become one of the largest custom cushion company in the world. Northcape is a leading brand in outdoor furnishing selling products through marketplaces like Wayfair, Dealers and B2B customers
While our advanced technology and customer focus are certainly our salient attributes, our true strength comes from our team of 1500+ people. Having scaled to USD 150mn in 13 years, we believe we are at the cusp of exponential and disruptive growth ahead.
We are seeking a highly experienced Senior AI Developer with deep expertise in Agentic AI systems to design, build, and scale autonomous AI agents capable of reasoning, planning, tool usage, multi-step execution, and real-world decision-making.
This role is not limited to prompt engineering. We are looking for someone who has built production-grade, multi-agent systems that integrate LLMs, vector databases, APIs, orchestration layers, and observability frameworks.
You will play a critical role in transforming our platform into an AI-First architecture by embedding intelligent agents across workflows, automation systems, and customer-facing applications.
Key Responsibilities
Agentic AI Architecture
- Design and implement autonomous AI agents capable of:
- Multi-step reasoning and task decomposition
- Tool usage (APIs, databases, search, calculators, external systems)
- Memory management (short-term + long-term)
- Self-reflection and corrective loops
- Architect multi-agent collaboration systems (Planner → Executor → Critic models)
- Implement ReAct, Tree-of-Thought, or other reasoning frameworks
System Engineering & Integration
- Integrate LLMs (OpenAI, Anthropic, open-source models) into scalable backend systems
- Build RAG pipelines using vector databases (Pinecone, Weaviate, OpenSearch, FAISS, etc.)
- Develop tool invocation frameworks and function-calling pipelines
- Optimize token usage, latency, and cost at scale
- Implement streaming architectures for real-time AI responses
Production-Grade Engineering
- Deploy AI systems on Kubernetes / cloud infrastructure (AWS / Azure / GCP)
- Build observability for:
- Prompt performance
- Hallucination tracking
- Agent failure modes
- Token cost monitoring
- Implement guardrails, safety layers, and compliance filters
- Design fallback and retry strategies
Advanced Capabilities (Preferred)
- Fine-tuning or LoRA adaptation of LLMs
- Building evaluation frameworks (LLM eval pipelines, scoring agents)
- Reinforcement Learning from Human Feedback (RLHF) familiarity
- Multi-modal AI (vision + text)
- AutoGPT / LangGraph / CrewAI / Semantic Kernel experience
Required Skills & Expertise
- Core AI Stack
- Python (mandatory), strong backend fundamentals
- LangChain / LangGraph / LlamaIndex
- Agent orchestration frameworks
- Prompt engineering at advanced level
- Vector databases & embeddings
- OpenAI / Anthropic APIs
- RAG architecture
Engineering & Infrastructure
- REST / GraphQL APIs
- Async programming
- Docker & Kubernetes
- CI/CD for ML pipelines
- Redis / Kafka (for orchestration)
- Strong system design capability
Architecture Knowledge
- Memory architectures (episodic, semantic)
- Tool calling frameworks
- Chain-of-thought optimization
- Evaluation & benchmarking strategies
- Distributed AI systems
What We’re Looking For
- Someone who has built real production AI agents, not just demos
- Ability to think architecturally (cost, scale, latency, security)
- Deep understanding of LLM limitations and mitigation strategies
- Strong debugging ability in complex reasoning chains
- Passion for building autonomous systems that replace manual workflows
Nice to Have
- Experience building AI copilots
- Experience in ecommerce / fintech / automation platforms
- Research background in multi-agent systems
- Published papers or open-source contributions
- Experience handling 1M+ AI requests/day systems
Impact of This Role
You will:
- Architect AI agents that automate complex workflows
- Build multi-agent AI systems that drive productivity gains
- Reduce operational cost through intelligent automation
- Lay the foundation for an AI-First enterprise platform
KPIs for Success
- % automation achieved through AI agents
- Reduction in manual intervention
- AI response latency under defined SLA
- Token cost efficiency improvement
- Agent accuracy and task completion rate