A Professional-Grade Program for Working Professionals.
Accelerate your career without pausing your work or studies. This 12-week Specialization is designed for software engineers, systems architects, and researchers who want to master the low-level architectural secrets behind the AI revolution. We focus on “Performance Engineering”—where every millisecond counts.
📅 Batch Starts: March 16th, 2026
✅ ARE YOU THE RIGHT FIT? (SELF-ASSESSMENT)
This is a high-intensity technical program. To maintain pace, we do not provide basic programming support. Ensure you meet the following:
- Target Audience: Software Engineers, Backend Developers, AI/ML Infrastructure Engineers, Recent Graduates , and Researchers.
- 🔴 CRITICAL PREREQUISITE: Proficiency in C Programming. You must be comfortable with Pointers, Dynamic Memory Allocation (malloc/free), and basic Data Structures.
- Format: Optimized for those with limited daily bandwidth. 60 hours of deep learning spread over 12 weeks.
- Commitment: Can you dedicate 5 hours every week for 12 weeks? This is a high-intensity hands-on programming course. 80% attendance is mandatory for receiving the course completion certificate.
🚀 The Specialization Edge
- Professional-Level Mastery: Transition from high-level abstractions to writing raw, high-performance kernels.
- Long-Term Skill Retention: A 12-week steady-paced deep dive allows for better absorption and application of complex hardware concepts.
- Industry-Standard Tooling: Direct exposure to professional profilers and debuggers from Day 1 to visualize hardware execution.
- SIMT Mastery: Master the Single Instruction, Multiple Threads paradigm—the backbone of AI and modern Supercomputing.
📚 Core Curriculum (Topics)
- Architectural Foundations: Decoding GPU vs. CPU execution flows and hardware-level parallelism.
- Kernel Development in C: Implementation of massively parallel functions.
- The Memory Challenge: Strategic use of Global, Shared, and Constant memory to bypass hardware latency.
- Performance Analysis: Real-time visualization of hardware bottlenecks using industry-standard tools.
- Concurrency Engineering: Utilizing Streams to overlap compute and data movement.
🛠️ Industry Toolkit
Master the tools used by performance engineers to visualize, debug, and optimize code:
- NVIDIA® Nsight™ Systems: For system-wide performance analysis and visualizing CPU-GPU interactions.
- NVIDIA® Nsight™ Compute: For interactive kernel profiling and deep-dive hardware metric analysis.
- cuda-gdb: Professional-grade debugging for parallel threads.
- cuda-memcheck / Compute Sanitizer: For detecting memory leaks, misaligned accesses, and race conditions.
- NVCC Compiler: Mastering compilation flags for architecture-specific optimization.
⏰ Professional Schedule (60 hours total)
Designed for working professionals. Total 5 hours per week:
- Live Online Interactive Classes (3 Hrs/Week):
- Mondays: 07:30 PM – 09:00 PM
- Thursdays: 07:30 PM – 09:00 PM
- Autonomous Lab Work (2 Hrs/Week): Flexible timing for assignments and deep practice.
⚠️ Hardware Requirement: Participants must arrange their own GPU access (Local GPU or Cloud platforms like Google Colab). GPU access is not provided.
🎙️ About The Speaker
Dr. Mandar Gurav is a Parallel Programmer who enjoys helping people accelerate their applications on CPU and GPU platforms. He previously worked with IITBombay, Nvidia, Intel, Centre for Development of Advanced Computing (C-DAC) and MulticoreWare before founding Siddhivinayak Advanced Computing Labs Pvt Ltd. Over the last 16 years, he has delivered a number of CPU/GPU parallelization projects in the following domains – Atmospheric Modelling, Computational Fluid Dynamics, Circuit Simulation, Robotics, Haptics, Video processing. As a part of professional service, he is involved in conducting training programs, delivering sessions on his areas of expertise in government research laboratories, industry and educational institutes (IITs, NITs, Engineering Institutes etc).
He holds a PhD from Indian Institute of Technology Bombay (IITBombay) and Bachelor of Engineering (Computer Science and Engineering) from Walchand College of Engineering, Sangli.
💰 Inaugural Batch Offer (50% Discount)
Take advantage of our subsidized launch pricing for this first “Industry-Ready” batch:
- Working Professionals: ₹8,000/- only (Original Price: ₹16,000/-)
- Early Bird Discount: ₹6,000/- only (Register before 10th March)
Note: This introductory pricing is subsidized to foster a community of high-performance developers.
⚠️ Enrollment Notice
Capacity: Strictly capped at 30 Seats. First-come, first-served for those who meet the prerequisites.
Registration (Razorpay/Payment Gateway Link): https://rzp.io/rzp/hpGNNsCS
Contact: contact@svacl.com / +91 9373881607
Stop Writing Slow Code. Start Accelerating.Note: This program utilizes the NVIDIA® CUDA™ platform. CUDA, Nsight, and NVIDIA are trademarks and/or registered trademarks of NVIDIA Corporation. All product names, trademarks, and registered trademarks mentioned herein are the property of their respective owners and are used solely for educational and informational purposes, with no implication of endorsement or affiliation.