About the Role

We’re hiring an experienced AI/ML Engineer to lead the development of a high-performance, multi-agent AI system using AutoGen, GPT-4, and a tightly integrated set of generative and data intelligence APIs.

You’ll be building the system’s core intelligence layer — reasoning agents, tool integrations, knowledge retrieval, and multimodal generation — in a modular, API-first environment. This is a deep engineering role with full ownership of the agent orchestration logic, generative pipeline, and backend interface.

Responsibilities

Core AI & Agentic System Development

• Design, develop, and maintain multi-agent reasoning workflows using AutoGen
• Build and configure agent roles (planner, researcher, generator, critic, etc.) with task-specific prompts and tool access
• Enable agent communication, memory sharing, reflection loops, and retry logic

LLM Integration & Prompt Engineering

• Integrate OpenAI GPT-4 Turbo via API with proper system message configuration, role memory, and adaptive chaining
• Handle token optimization, dynamic context management, and API response control
• Implement prompt templates for reasoning, estimation, generation, and critique agents

External API Integration

• Integrate and manage the following tools and APIs:
• Serper.dev or Brave API for real-time search and data retrieval
• Galileo AI or Uizard APIs for UI mockup generation
• Mermaid.js (rendered via LLM) for architecture and flow diagrams
• Runway ML Gen-2 API for generating concept/demo videos
• ElevenLabs API for voice narration (optional integration)

Memory & Retrieval Engineering

• Implement semantic memory using ChromaDB or Pinecone
• Build document indexing, RAG pipelines, and long-context memory injection for agents
• Handle session management, version tracking, and memory-based prompt enrichment

Backend Development & Deployment

• Build modular backend endpoints using FastAPI for each agent or task chain
• Manage secure, scalable, and containerized deployment using Docker
• Set up CI/CD hooks and infrastructure for internal agent testing, logs, and caching

Required Skills

• Strong experience with Python, LLM APIs, and generative AI pipelines
• Deep understanding of AutoGen, agent orchestration, and LLM-driven reasoning systems
• Experience with prompt chaining, reflection logic, and tool-augmented LLM workflows
• Proficiency in integrating REST APIs and structuring external tool pipelines
• Familiarity with vector databases: ChromaDB or Pinecone
• Strong backend development skills using FastAPI, with production-ready code delivery

More the Merrier (Not Required)

• Familiarity with diagram generation (Mermaid, Graphviz, or similar)
• Experience using video or voice generation tools via API
• Understanding of LLM cost/latency optimization
• Exposure to multi-agent evaluation, self-critique loops, or scalable AI architecture