AI/ML Engineer (DLMs, Embeddings, Fine-Tuning)

Company Overview for Client

On behalf of a stealth-stage DeepTech AI company. This role is being recruited for a client that is transforming a major, multibillion dollar industry via proprietary AI.

Company Summary

We are representing a highly ambitious stealth DeepTech AI company operating across multiple international markets. Their core strength is a proprietary Deep Learning Model (DLM), fine-tuned on massive, domain-specific data to deliver performance superior to general-purpose foundation models. The company focuses on solving critical operational inefficiencies by providing AI-first automation, secure multi-platform tooling, and multilingual support. They are scaling rapidly toward significant ARR targets with strong unit economics and are positioned to capture a large market opportunity.

AI/ML Engineer (DLMs, Embeddings, Fine-Tuning)

Job Title: AI/ML Engineer

Employment Type: Full-Time

Location: Hyderabad, India

Experience: 5+ Years in senior role.

Role Summary

You will be the core technical driver behind the clients competitive moat: the proprietary Deep Learning Model (DLM). This is a hands-on, highly technical role focused on achieving and maintaining domain-specific accuracy and cost-efficiency that generic models cannot replicate. You will own the end-to-end lifecycle of models from data ingestion and fine-tuning to deployment and optimization ensuring the platform delivers industry-leading accuracy (targeting 95% on complex domain benchmarks) while operating at a highly competitive cost advantage.

Key Responsibilities

• Proprietary Model Fine-Tuning: Lead fine-tuning and customization of Large Language Models (LLMs) or similar Deep Learning Models using techniques such as LoRA/PEFT for domain-specific performance.
• Design and manage high-volume data ingestion and cleaning pipelines; implement automated QA checks and coordinate expert corpus review.
• Develop, manage, and optimize embeddings generation and vector search workflows; integrate with vector databases (e.g., Qdrant or similar) to enable accurate Retrieval-Augmented Generation (RAG).
• Cost & Performance Moat: Optimize model inference for maximum throughput and minimal cost-per-query, specifically targeting operations that are up to 1000x cheaper than large commercial general-purpose models using techniques like distillation and quantization .
• Implement and refine core AI algorithms for specialized tasks such as predictive insights and automated content extraction.
• Collaborate on deploying and monitoring AI microservices in production with a focus on scalability, reliability, and observability.

Required Technical Skills

• 5+ years of experience in AI/ML with deep specialization in LLMs, NLP, and fine-tuning techniques.
• Persona: Must be a highly autonomous, senior individual contributor with a commitment to engineering and optimizing proprietary, deep-tech IP.
• Expert-level Python and strong proficiency with PyTorch or TensorFlow.
• Practical experience with vector databases, embeddings pipelines, and RAG architectures.
• Demonstrated ability to optimize models for production deployment including GPU resource management, quantization, and distillation.
• Strong knowledge of data cleaning methodologies and advanced ML algorithms.

What We Offer

• We offer a competitive salary aligned with the Hyderabad startup market.
• Continuous learning support including conference allowance and learning resources.
• Collaborative, lean team culture where decisions move fast and contributions are visible.
• Opportunity for rapid career growth as the company scales and secures funding.

How to Apply

Please apply via LinkedIn Easy Apply or email with the following:

1. Updated CV highlighting relevant LLM, embeddings, and fine-tuning work.
2. Short cover note (23 paragraphs) describing a recent project where you improved model accuracy or cost-efficiency; include the approach, tools, and measurable outcome.
3. Links to project artifacts (GitHub, Colab notebooks, papers, demo videos) if available.

Candidates who pass initial screening will be asked to answer 23 short technical questions and provide a concise case write-up or code sample demonstrating relevant experience.

Join Us and Make an Impact

By joining this team, you will contribute directly to a platform driving significant efficiency gains and reducing friction across large industry workflows. The client seeks professionals ready to commit, innovate, and scale as they pursue market leadership and industry transformation.

#DeepTech #AIMLEngineer #LLM #Quantization #RAG #HyderabadJobs #Startup