Senior AI/ML Engineer

Recro; Bengaluru; On-site; INR 1,800,000 - 3,200,000; Full time

Job summary

Recro is seeking an AI Engineer to own production-grade AI systems end-to-end, from LLM applications to model serving and monitoring.

The ideal candidate will have hands-on experience in Agentic AI, LLM orchestration, self-hosted models, and production ML engineering, with a strong grounding in Python, PyTorch, and vector databases.

Qualifications

• Education: IIT/NIT/BITS required.
• 4–7 years of experience in AI/ML production systems.
• Strong hands-on in Agentic AI and LLM systems.

Responsibilities

• Build AI agents, LLM workflows, and orchestration systems.
• Develop RAG pipelines, memory systems, and retrieval workflows.
• Work with LLM gateways and agent frameworks.
• Improve reliability and performance of LLM applications.
• Deploy self-hosted AI models and optimize latency.

Skills

Agentic AI

LLM / Generative AI

LLM Orchestration

LLM Gateway

RAG

Vector Databases

vLLM / Triton

Python

PyTorch

MLOps

Education

IIT/NIT/BITS degree

Tools

Docker

Kubernetes

Cloud tooling

Job description

AI Engineer – Agentic AI & Production ML

Experience 4–7 Years

Education: Only IIT OR NIT OR BITS

Role Overview

We are looking for an AI Engineer to build and own production-grade AI systems end-to-end — from LLM applications and agentic workflows to model serving, evaluation, and monitoring.

The role requires strong hands-on experience in Agentic AI, LLM orchestration, self-hosted models, and production ML engineering.

Responsibilities

Agentic AI & LLM Systems (Must Have)

• Build AI agents, LLM workflows, and orchestration systems.
• Develop RAG pipelines, memory systems, and retrieval workflows.
• Work with LLM gateways and agent frameworks.
• Improve reliability and performance of LLM applications.

Model Serving & Infrastructure (Must Have)

• Deploy and optimize self-hosted AI models.
• Experience with vLLM / Triton or similar serving frameworks.
• Optimize latency, scalability, and inference cost.
• Work with Docker, Kubernetes, and cloud infrastructure.
• Build production AI systems using Python.
• Work with PyTorch, Hugging Face, Transformers, and vector databases.

Evaluation & Observability

• Build evaluation frameworks and benchmark datasets.
• Track AI quality, latency, cost, and reliability metrics.
• Set up monitoring, dashboards, and alerts for production AI systems.

Speech AI (Good to Have)

• Experience with STT, ASR, speaker diarization, or voice AI systems.

Required Skills

Must Have

• Agentic AI
• LLM / Generative AI
• LLM Orchestration
• LLM Gateway
• RAG
• Vector Databases
• vLLM / Triton
• Python
• PyTorch
• MLOps

Good to Have

• RLHF
• AI Evaluation
• Open-source AI contributions

Ideal Candidate Profile

• Has shipped AI products into production.
• Strong understanding of LLM systems and AI architecture.
• Can own complete AI systems, not just model tuning.
• Demonstrates strong engineering fundamentals and problem-solving ability.

Hiring benchmark: Candidates should demonstrate strong Agentic AI depth, production ML experience, and ownership of scalable AI systems.