Foundational Model Engineer

Company Description

is at the forefront of engineering India's Scientific AI Future. Selected by IndiaAI MeitY, we are building the nation's Sovereign Scientific Foundation Model to advance India's leadership in AI. Unlike traditional LLMs, our models learn from equations, simulations, and the fundamental laws of engineering, producing physics-consistent outputs. By combining physics, mathematics, and data, we are revolutionizing AI to reason from first principles, thereby significantly compressing R&D cycles. Founded by IISc faculty and incubated at India's premier research institution, we are dedicated to creating a sovereign scientific AI stack with a focus on compute, datasets, benchmarks, and safety frameworks.

Role Description

This is a full-time on-site role for a Foundational Model Engineer based in Bengaluru. The Foundational Model Engineer will be responsible for the development and optimization of Scientific Foundation Models, ensuring they adhere to principles of physics and engineering. Daily tasks include algorithm development, data processing, simulation, and collaboration with cross-functional teams to integrate AI models into various engineering applications. Additionally, the engineer will contribute to refining compute infrastructure and safety frameworks essential for the AI stack.

Qualifications

• Algorithm Development, Data Processing, and Simulation skills
• Experience with AI Model Development and Optimization
• Proficiency in Physics and Engineering Principles
• Strong collaboration and teamwork skills in cross-functional environments
• Excellent problem-solving and analytical skills
• Master's or Ph.D. in Computer Science, Engineering, or related field
• Experience in scientific research and AI in strategic sectors is a plus

As a Foundation Model Engineer at ZenteiQ, you will lead the development of large-scale Scientific Foundation Models that integrate deep learning with physical and scientific principles. This role blends AI research, systems engineering, and architectural innovation to create models capable of reasoning, prediction, and scientific understanding.

Your work will directly contribute to ZenteiQ's mission to build sovereign, ethical, and high-performance scientific AI systems that transform industries such as aerospace, materials science, and energy.

Key Roles & Responsibilities:

• Model Architecture: Design and implement scalable foundation model architectures (10B–100B+ parameters) using transformer and hybrid attention mechanisms.
• Distributed Training: Build high-efficiency training pipelines with FSDP, DeepSpeed, and Megatron-LM to optimize multi-node GPU workloads.
• Multimodal Integration: Develop cross-modal models combining text, image, simulation, and sensor data for scientific applications.
• Performance Optimization: Enhance inference and training performance using quantization, KV-cache management, mixed precision, and CUDA kernel optimization.
• Research and Experimentation: Conduct ablation studies, evaluate new architectures (MoE, GQA, MQA), and publish results in internal and external forums.
• Collaboration: Work closely with the data, infrastructure, and HPC teams to ensure scalable deployment and compliance with security standards.
• Mentorship: Guide junior engineers in model design, training efficiency, and research best practices.

Project Goals:

• Develop and deploy ZenteiQ's Scientific Foundation Model (SFM) capable of reasoning over scientific data and equations.
• Achieve breakthroughs in physics-informed deep learning, integrating domain constraints with AI-driven learning.
• Advance India's sovereign AI initiative by building a national-scale scientific AI system optimised for precision, interpretability, and scalability.
• Contribute to open research publications, ethical AI practices, and reproducible model design frameworks.

Requirements

Required Skills & Qualifications:

• Master's or Ph.D. in Computer Science, AI/ML, or related field.
• Expertise in transformer-based architectures (GPT, LLaMA, T5, etc.)
• Proficiency in PyTorch, CUDA, and distributed training frameworks (DeepSpeed, FSDP, Megatron-LM)
• Experience with HPC environments, multi-GPU scaling, and parallelization strategies (tensor, pipeline, data)
• Strong understanding of numerical methods, scientific computing, or physics-informed ML
• Proven track record in large-model research, optimization, or deployment
• Publications or open-source contributions in foundation models preferred.

Tech Stack:

PyTorch
• CUDA
• Flash Attention
• DeepSpeed
• FSDP
• Megatron-LM
• JAX
• Triton
• MoE
• RoPE / ALiBi
• xFormers
• HPC
• Linux

Benefits

• Work on cutting-edge national-scale AI systems.
• Access to H100/H200 GPU clusters and next-gen compute infrastructure.
• Competitive compensation, performance bonuses, and equity options.
• Opportunity to publish and present at global AI conferences.
• A collaborative, ethical, and innovation-driven work environment.

Why Join Us

• Join  — a DeepTech startup founded by IISc Bangalore faculty, pioneering AI at the intersection of science & engineering and human intelligence.
• IndiaAI  Impact: Selected by the IndiaAI Mission, MeitY, to build India's Scientific Foundation Model (SFM) — advancing the nation's sovereign AI capabilities. Watch Announcement
• Human-Centered AI: Work on next-generation personalized AI companions that adapt, reason, and evolve with users — redefining how people engage with intelligence.
• Cutting-Edge R&D Environment: Collaborate with scientists, engineers, and researchers on LLMs, SciML, and adaptive systems — bridging research and real-world innovation.
• Purpose with Ownership: Be part of a culture that values integrity, autonomy, and bold experimentation — where every idea can shape the future of AI for both science and society.

Pls apply @ Jobs at ZenteiQ Aitech Innovations Private Limited