Overview

We are looking for a production-minded AI/ML Engineer to design, build, train, and deploy the machine learning systems that power our real-time copilot.

This is a full-stack ML role spanning the entire lifecycle — from data pipelines and model training to deployment, optimization, and monitoring in production.

You will work at the intersection of LLMs, speech AI, retrieval systems, and real-time inference, building systems that must be fast, reliable, and scalable for millions of users.

Responsibilities

Key Responsibilities

1. Model Development & Training

Build, train, and fine-tune ML models including LLMs, NLP systems, and speech-to-text models
Apply modern techniques like LoRA, QLoRA, DPO, RLHF, and adapters
Design high-quality training datasets through cleaning, annotation, and augmentation
Run experiments, ablations, and performance evaluations to improve model quality

2. Real-Time Model Deployment

Build low-latency inference pipelines for real-time AI responses
Optimize performance using streaming, caching, batching, and quantization
Deploy models using Docker, Kubernetes, and CI/CD pipelines
Integrate multiple LLM providers with routing, failover, and orchestration systems

3. RAG & AI Systems

Build retrieval-augmented generation (RAG) pipelines using embeddings and vector databases
Develop semantic search, summarization, classification, and NLP features
Design prompt systems, tool calling, and multi-turn conversational flows
Build agentic workflows with function calling and multi-step reasoning

4. Evaluation & Monitoring

Create automated and human evaluation systems for model quality
Track latency, hallucinations, accuracy, cost, and user satisfaction metrics
Implement drift detection and regression monitoring
Optimize token usage and inference costs across models

5. Data & Feature Engineering

Build scalable data pipelines for training and production use
Design feature extraction systems from user interactions and logs
Implement data versioning, lineage tracking, and experiment tracking
Collaborate with analytics and engineering teams on data infrastructure

6. Safety & Responsible AI

Implement guardrails, filters, and safety validation systems
Evaluate models for bias, fairness, and robustness
Ensure privacy-first design and secure handling of user data
Support responsible AI practices across the ML lifecycle

Qualifications

Required Qualifications
Experience
3+ years in ML engineering or AI system development
Proven experience shipping production AI/ML systems end-to-end
Strong cross-functional collaboration experience
Startup or fast-paced environment experience preferred
Technical Skills
Strong Python and ML frameworks (PyTorch, TensorFlow, JAX)
Deep understanding of transformers, deep learning, and optimization
Experience with LLM APIs (OpenAI, Anthropic, open-source models)
Hands-on RAG and vector database experience
Docker, Kubernetes, FastAPI, CI/CD for ML systems
Cloud platforms (AWS, GCP, or Azure)