AI Developer (LLM Specialist)

• *AI Engineer**

• *Location:** On\-Site — Delhi, India **Type:** Full\-time \| Immediate Joiners Preferred

We're building a next\-gen AI product that personalizes conversations at scale using Large Language Models. We're looking for a hands\-on **AI Engineer** who can own LLM systems end\-to\-end — from designing agentic pipelines and writing the Python backend to engineering the prompts and evals that make the output reliable.

This is a build\-from\-scratch role. You should be comfortable across the whole stack: orchestration, retrieval, serving, and — critically — the prompt layer where most of the product's quality is won or lost.

• *What You'll Do**

• Design and build agentic / multi\-step LLM pipelines using **LangGraph** (or LangChain)
• Write production Python backends — FastAPI services, REST APIs, background workers
• Build and optimize RAG pipelines: retrieval, grounding, citation, context shaping
• **Own prompt engineering as a core discipline** — design, test, and version production prompts (system prompts, few\-shot, chain\-of\-thought, structured JSON/XML outputs) and build the evals that keep them reliable
• Systematically reduce hallucination and improve grounded, citation\-backed outputs
• Integrate vLLM, OpenAI, Anthropic, or open\-source models for fast inference
• Containerize and deploy services with Docker
• Optimize latency, token usage, cost, and response quality
• Stay current with Generative AI \& MLOps developments

• *Key Skills We're Looking For**

• **Strong prompt engineering** — deep, hands\-on command of few\-shot, chain\-of\-thought, system\-prompt design, structured outputs, prompt versioning, and an eval\-driven approach (promptfoo, LangSmith, or custom). This is a must\-have, not a bonus.
• **Strong Python** — clean, production\-grade, not just notebooks
• **LangGraph / LangChain** — hands\-on experience building agentic or multi\-node pipelines
• **LLM APIs** — OpenAI, Anthropic, Hugging Face, Cohere
• **RAG** — vector databases (Pinecone, FAISS, Qdrant), retrieval and grounding
• **FastAPI \+ REST API** development
• **Docker** and containerized deployments
• Bonus: vLLM or other model\-serving frameworks, streaming (SSE/WebSockets), MLOps tooling

• *Ideal Profile**

• 2–4 years building AI/ML products (strong 1\-year candidates with an exceptional portfolio considered)
• Proven projects or GitHub showing real LLM systems — agents, RAG, or chatbots that actually shipped
• A portfolio that demonstrates prompt craft: prompts, eval suites, or measurable quality improvements
• Curious, self\-driven, and fast at turning ideas into working systems
• Excited to build AI that solves real business problems

• *Why Join Us?**

• Build from scratch in a high\-impact B2B AI product
• Work directly with leadership \& product strategy
• Lean, execution\-focused founding team
• Flexible setup, high ownership, fast iterations

• *How to Apply** Send your resume and GitHub/portfolio to **akshit.kaushik@originbluy.com** Subject line: **AI Engineer Application – \[Your Name]**

Job Types: Full\-time, Permanent

Pay: ₹20,000\.00 \- ₹25,000\.00 per month

Benefits

• Commuter assistance
• Food provided
• Health insurance
• Internet reimbursement

Application Question(s)

• minimum 2 year as an AI developer

Experience

• AI: 2 years (Required)
• Machine learning: 1 year (Required)

Work Location: In person