Principal AI Engineer (LLM Agents & Orchestration)
This job may have closed, so you can apply to similar jobs
6 live jobs like this one, posted recently - apply in one step.
Principal Engineer – GenAI & Agentic AIApply nowApplyThe Wells Fargo FoundationBengaluru, India · 5-7 yrs · Posted 3 days ago
Lead Speciality Software Engineer GenAI & Agentic AIApply nowApplyThe Wells Fargo FoundationBengaluru, India · 5-7 yrs · Posted 3 days ago- Senior Agentic AI Developer / Solution EngineerApply nowApplyPapigen Private LimitedChennai, Tamil Nadu · 8-10 yrs · Posted 2 days ago
AES - DE - Generative AI Prompt EngineersApply nowApplyZensar TechnologiesBengaluru / Bangalore, Karnataka · 8-10 yrs · Posted 2 days ago- AI Security Operations Engineer (Teradyne, India)Apply nowApplyTeradyneBengaluru / Bangalore, Karnataka · 5-9 yrs · Posted 12 days ago
Principal Software Engineer(Autodesk AI Assistant)Apply nowApplyAutodeskPune, Maharashtra · 10-12 yrs · Posted 20 days ago
An AI agent that applies to roles like this on employers' own sites. 25 applications free, no card needed.
Job Description
Job Title: Principal AI Engineer (LLM Agents & Orchestration)
Team: Chatly Engineering
Focus: Building Autonomous Super Agents
Location: On-Site — H-12, NSTP, Islamabad
About The Role
We are looking for a deep expert in Large Language Models (LLMs) to lead the architectural development of our new Super Agent within Chatly. You will move beyond simple chat interfaces to build autonomous agents capable of complex reasoning, tool usage, and seamless integration with external workflows.
Key Responsibilities
- Agent Architecture: Design and implement stateful agentic workflows (using frameworks like LangGraph or custom Python/TypeScript solutions) that can plan, execute, and self-correct
- Integration Ecosystem: Build the hands of the agent. Develop robust integration layers that allow the LLM to interact with our internal APIs, databases, and third-party tools reliably
- Latency & Reliability: Optimize inference pipelines for speed (streaming, token optimization) and reliability (handling hallucinations, structured output validation)
- Memory Systems: Architect advanced RAG (Retrieval-Augmented Generation) systems to give the Super Agent persistent memory and context awareness across sessions
- Evaluation & Observability: Establish a rigorous testing framework for non-deterministic model outputs to ensure the agent behaves as expected in production
Technical Requirements
- Core Stack: Deep proficiency in Python (or TypeScript) and the modern AI stack (LangChain, LlamaIndex, DSPy)
- Model Internals: Strong grasp of how to leverage specific model strengths (e.g., GPT-4o for reasoning, Haiku/Flash for speed) and experience with Function Calling/Tool Use
- Vector Search: Hands-on experience with vector databases (Pinecone, Milvus, Weaviate) and embedding strategies
- System Design: Experience designing event-driven architectures where agents respond to triggers, not just user prompts
Why Join ImagineArt
- Work on innovative AI products like ImagineArt, shaping the future of creativity and user interaction
- Be part of a passionate, fast-moving team that values innovation and data-driven decisions
- Competitive compensation and benefits
- A culture where learning, growth, and experimenting with new ideas are deeply encouraged
Company Culture
- Impact: Opportunity to work on innovative AI products like ImagineArt, shaping the future of creativity and user interaction
- Team: Be part of a passionate, fast-moving team that values innovation and data-driven decisions
- Growth: A culture where learning, growth, and experimenting with new ideas are deeply encouraged
Skills
PythonLarge Language ModelsAi EngineerAiIf a job posting appears fraudulent, asks for payment, contains misleading information, or violates our guidelines, please report it immediately. Our team will review it promptly, Jobaaj does not charge any fee from the applicants.
About Company
Similar Jobs
View allLooking for something similar?
New Data Science Analytics & Machine Learning roles in India.
One email a day. Unsubscribe any time.

