AI / ML Engineer

OptiKratos By Synoviq, Inc.; India; On-site; INR 1,200,000 - 2,400,000; Full time

Job summary

OptiKratos in India is seeking an experienced ML engineer to own the AI workstreams, shipping production features such as voice and video agents, RAG-powered retrieval, and CRM write-back.

You will work with LLM APIs, optional GPU inference, and the WhatsApp Graph API to build scalable agents for revenue teams.

Collaborate across data, platform, and product teams to drive latency-optimized, cost-aware deployments.

Qualifications

• 3+ years of ML engineering or applied AI development in production systems.
• Strong Python skills; experience with LangChain, LlamaIndex, or equivalent RAG frameworks.
• Familiarity with LLM APIs (OpenAI, Anthropic) and prompt engineering.
• Experience with vector databases (pgvector, Pinecone, Weaviate, or similar).
• Understanding of async Python and background task queues (Celery, Redis).

Responsibilities

• Design and ship AI agent workflows: voice/video agents, autonomous outreach, and lead qualification.
• Build and maintain RAG pipelines — document ingestion, embedding, retrieval, and response generation.
• Implement CRM write-back: parse AI conversation outputs and update CRM records via internal APIs.
• Integrate with Meta WhatsApp Graph API v26.0 for AI-powered messaging flows.
• Evaluate and benchmark LLMs (OpenAI, Anthropic, open-source) for latency, accuracy, and cost.
• Work with optional GPU inference infrastructure for self-hosted model serving.

Skills

Python

LangChain

LlamaIndex

RAG

LLM APIs

Prompt Engineering

Tools

pgvector

Pinecone

Weaviate

Redis

Celery

OpenAI

Anthropic

Job description

You will own the modules/ai workstream — shipping production AI features: voice and video agents, RAG-powered knowledge retrieval, CRM write-back, autonomous lead qualification, and sentiment analysis.

You will work with LLM APIs, optional GPU inference, and the WhatsApp Graph API v26.0 to build agents that operate at scale for revenue teams.

Responsibilities

• Design and ship AI agent workflows: voice/video agents, autonomous outreach, and lead qualification.
• Build and maintain RAG pipelines — document ingestion, embedding, retrieval, and response generation.
• Implement CRM write-back: parse AI conversation outputs and update CRM records via internal APIs.
• Integrate with Meta WhatsApp Graph API v26.0 for AI-powered messaging flows.
• Evaluate and benchmark LLMs (OpenAI, Anthropic, open-source) for latency, accuracy, and cost.
• Work with optional GPU inference infrastructure for self-hosted model serving.

Requirements

• 3+ years of ML engineering or applied AI development in production systems.
• Strong Python skills; experience with LangChain, LlamaIndex, or equivalent RAG frameworks.
• Familiarity with LLM APIs (OpenAI, Anthropic) and prompt engineering.
• Experience with vector databases (pgvector, Pinecone, Weaviate, or similar).
• Understanding of async Python and background task queues (Celery, Redis).