Welcome to my LinkedIn profile! I am Aman K Shihab, an Associate Engineer specializing in AI at Ignitarium Technology Solutions Pvt. Ltd. in Bangalore, India. With a Bachelor's degree in Computer Science and Engineering from Federal Institute Of Science And Technology, Kerala, I have developed a strong foundation in cutting-edge technologies and their practical applications.At Ignitarium, I have been instrumental in designing and implementing high-performance convolutional layer kernels for a massively parallel Edge AI inference chip. My work involves optimizing algorithms in Python and C++ to enhance memory utilization and processing speed, ensuring maximum performance and minimal latency in AI inference tasks. Collaborating with my team, I have fine-tuned memory management strategies and developed test suites for kernel performance evaluation and optimization.As an Undergraduate Researcher at the Cognitive Computing Research Center, I focused on improving the performance of pose detection deep learning models and developed a state-of-the-art model for named entity recognition in Malayalam. Also built an end-to-end speaker diarization and transcription pipeline using state-of-the-art machine learning techniques. I also deployed scalable FastAPI endpoints on GCP using Kubernetes, optimizing resource utilization for cost-effective operations. My technical skills include proficiency in Python, C, C++, SQL, JavaScript, HTML/CSS, and frameworks like PyTorch, Flask, Django, React, FastAPI, pandas, NumPy, Matplotlib, and scikit-learn. I am passionate about contributing to open-source projects, with notable contributions such as Named Entity Recognition In Malayalam and EfficientPose for pose estimation.I am also a co-author of a paper on Lithium-Ion Battery Modeling using Machine Learning techniques, which won the Best Paper award at ICSET 2022. My projects span across computer vision, deep learning, IoT, and software development, showcasing my versatility and commitment to innovation.Feel free to connect with me to discuss AI, machine learning, software development, or any exciting opportunities for collaboration!
-
EngineerIgnitariumKochi, Kl, In
-
Associate EngineerIgnitarium Sep 2023 - PresentBengaluru, Karnataka, IndiaDesigned and implemented high-performance convolutional layer kernels for a massively parallel Edge AI inference chip.Utilized Python and C++ to optimize algorithms for efficient memory utilization and processing speed.Gained a deep understanding of chip architecture to maximize performance and minimize latency in AI inference tasks.Collaborated with team members to fine-tune memory management strategies, ensuring optimal use of available resources.Developed and implemented test suites for kernel performance evaluation and optimization.Constantly researched and implement cutting-edge techniques to improve kernel efficiency and overall chip performance.
-
Undergraduate ResearcherCognitive Computing Research Center (Ccrc) Sep 2022 - Jun 2023Kerala, IndiaTrained a cutting edge deep learning model based on Transformer for Named Entity Recognition in Malayalam.Achieved an F1 Score of 0.98 on the test set.Dealt with imbalanced classes using weighted loss functions.Fine-tuned a pose detection vision model to improve it's accuracy by 2% and scaled down the network to run on consumer CPU'sModified the Tensorflow model playground to generate the code for the visualized neural network.
Aman K Shihab Education Details
-
Federal Institute Of Science And TechnologyComputer Science
Frequently Asked Questions about Aman K Shihab
What company does Aman K Shihab work for?
Aman K Shihab works for Ignitarium
What is Aman K Shihab's role at the current company?
Aman K Shihab's current role is Engineer.
What schools did Aman K Shihab attend?
Aman K Shihab attended Federal Institute Of Science And Technology.
Who are Aman K Shihab's colleagues?
Aman K Shihab's colleagues are Vishnu Krishnan, Shabeena K M, Sourav Raj K K, Krishna Reddy, Gowtham Unni T, Bharath Reddy, Rohish Bhaskaran.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial