Apply to job

Soothsayer Analytics is a global AI and Data Science consultancy headquartered in Detroit, with a thriving delivery center in Hyderabad. We design and deploy end-to-end custom Machine Learning solutions spanning predictive analytics, optimization, NLP, and GenAI that help leading enterprises forecast, automate, and gain a competitive edge.

Join us to tackle high-impact, cross-industry projects where your ideas move rapidly from concept to production, shaping the future of data-driven decision-making.

We seek a Senior AI Scientist with strong ML fundamentals and data engineering expertise to lead the development of scalable AI/LLM solutions. You will design, fine-tune, and deploy models (e.g., LLMs, RAG architectures) while ensuring robust data pipelines and MLOps practices.

Key Responsibilities

1. AI/LLM Development:
• Fine-tune and optimize LLMs (e.g., GPT, Llama) and traditional ML models for production.
• Implement retrieval-augmented generation (RAG), vector databases, and orchestration tools (e.g., LangChain).
2. Data Engineering:
• Build scalable data pipelines for unstructured/text data (e.g., Spark, Kafka, Airflow).
• Optimize storage/retrieval for embeddings (e.g., pgvector, Pinecone).
3. MLOps & Deployment:
• Containerize models (Docker) and deploy on cloud (AWS/Azure/GCP) using Kubernetes.
• Design CI/CD pipelines for LLM workflows (experiment tracking, monitoring).
4. Collaboration:
• Work with DevOps to optimize latency/cost trade-offs for LLM APIs.
• Mentor junior team members on ML engineering best practices.

Required Skills & Qualifications

• Education: MS/PhD in CS/AI/Data Science (or equivalent experience).
• Experience: 6+ years in ML + data engineering, with 2+ years in LLM/GenAI projects.

Skills Matrix

Candidates must submit a detailed resume and fill out the following matrix:

Instructions for Candidates:

1. Provide a detailed resume highlighting projects related to LLMs, data engineering, and MLOps.
2. Fill out the matrix above with accurate dates, experience duration, and self-ratings.