I am a Ph.D. candidate in the Electrical and Computer Engineering Department at the University of Wisconsin-Madison, working with Prof. Umit Ogras. My research focuses on efficient AI systems, including hybrid Mamba-Transformer and Mixture-of-Experts (MoE) inference, heterogeneous accelerator and chiplet architectures, thermal modeling, and thermal-aware workload scheduling.

During Summer 2026, I worked as a Server Research Intern at ByteDance on LLM inference for heterogeneous racks. Previously, I was an Architecture Research Intern at Ayar Labs and worked at Samsung Electronics and Imagination Technologies on system architecture, 3D ICs, RTL design, and design automation. I completed my undergraduate studies at the Indian Institute of Technology (IIT) Jodhpur, where I collaborated with Dr. Neeraj Goel and Dr. Dip Sankar Banerjee.

In case you want to know more about my research or want to collaborate with me, please feel free to send me an email at ahkanani@wisc.edu.

CV | Prelim Slides | Abstract

Updates:

[Aug-2026]: Three papers have been accepted at IEEE/ACM Embedded Systems Week (ESWEEK) 2026: “APEX: Adaptive Expert Prefetching for Memory-Efficient Edge MoE Inference”, “HYDRA: A Heterogeneous Chiplet DSE Framework for Serving Dynamic Hybrid LLM Workloads”, and “ReVolt: Power Delivery Network-Aware Voltage Droop Control for 2.5D PIM Chiplet Architectures”.

[Jun-2026]: Started a Server Research Internship at ByteDance, working on MoE LLM inference with heterogeneous GPU and accelerator racks.

[Jun-2026]: Our paper “MFIT: Multi-Fidelity Thermal Modeling for 2.5D and 3D Multi-Chiplet Architectures” received the ACM TODAES Best Paper Award, presented at the 63rd DAC. Read the UW-Madison feature

[Jun-2026]: I won First Place at the DAC Ph.D. Forum for my dissertation research, “Cross-Layer Design of Heterogeneous Multi-Chiplet Systems for Efficient and Scalable AI Inference.” Poster.

[Mar-2026]: Our U.S. patent application, “Runtime Monitoring of Machine Learning-Based Scheduling Algorithms Toward Robust Domain-Specific Systems-on-Chip”, was published as US20260093522A1.

[Mar-2026]: Two papers have been accepted at Design Automation Conference (DAC) 2026: “DUET: Disaggregated Hybrid Mamba-Transformer LLMs with Prefill and Decode-Specific Packages” and “LEXI: Lossless Exponent Coding for Efficient Inter-Chiplet Communication in Hybrid LLMs”.

[Feb-2026]: MFIT was selected as an Editor’s Pick by ACM Transactions on Design Automation of Electronic Systems (TODAES).

[Oct-2025]: Our paper “CHIPSIM: A Co-Simulation Framework for Deep Learning on Chiplet-Based Systems” has been accepted at the IEEE Open Journal of the Solid-State Circuits Society (OJ-SSCS) Special Issue on Chiplet Interconnect and Architectures.

[Oct-2025]: Passed my Ph.D. Preliminary Exam and am now a Doctoral Candidate, halfway point of Ph.D. journey!

[Oct-2025]: Our paper “HeMu: Energy-Efficient DNN Inferencing via Heterogeneous-Multi-Chiplet Architectures” has been accepted at the IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD).

[Oct-2025]: Awarded an NSF travel grant to present “THERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures” in Taipei!

[Aug-2025]: Our paper “MFIT: Multi-Fidelity Thermal Modeling” has been accepted at the ACM Transactions on Design Automation of Electronic Systems (TODAES).

[July-2025]: Gave an invited talk at the Dutt research group on “MFIT: Multi-Fidelity Thermal Modeling”.

[Jun-2025]: Two papers have been accepted at ESWEEK 2025: “THERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures” and “eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing”.

[May-2025]: Started my summer internship at Ayar Labs, working on AI/ML network topology and performance analysis.

[Feb-2025]: Presented “MFIT: Multi-Fidelity Thermal Modeling” at the Wisconsin CHIPS Meeting.

[Oct-2024]: Completed my Research Readiness/Qual. Exam.

[Oct-2024]: Gave a talk on “Runtime Monitoring of ML-based Scheduling Algorithms Toward Robust Domain-Specific SoCs” at the Computer Engineering Department Seminar Series.

[Jun-2024]: I have been selected as a DAC Young Fellow to attend the 61st Design Automation Conference (DAC) in San Francisco, CA.

[May-2024]: Paper on “Runtime Monitoring of ML-based Scheduling Algorithms Toward Robust Domain-Specific SoCs” has been accepted at ESWEEK 2024.

[Mar-2024]: Paper on “Thermal Modeling and Management Challenges in Heterogenous Integration: 2.5 D Chiplet Platforms and Beyond” has been invited for a special session at VTS 2024.

[Nov-2023]: Our paper “Ellora: Exploring Low-Power OFDM-based Radar Processors using Approximate Computing” has been accepted at LASCAS 2024.

[Aug-2023]: I started my Ph.D. studies at the University of Wisconsin-Madison.

[Apr-2023]: I was awarded the Employee of the Month at Samsung Electronics.

[Jul-2022]: Started working at Samsung Electronics as a System Architecture and RTL Design Engineer.

[Nov-2021]: Our paper “Approximating Communication Systems: Reality or Fantasy?” has been accepted at HiPC SRS 2021 for a poster presentation.

[Aug-2021]: Our paper “LightFPGA: A Lightweight Approximate FPGA for Energy-Efficient Approximate Computing” has been accepted at VDAT 2021.

[Jul-2021]: Started working at Imagination Technologies as a Hardware Engineer.

[Jul-2021]: Our paper “ApproxBioWear: An Approximate Computing Framework for Energy-Efficient Wearable Biomedical Devices” has been accepted at EMBC 2021.

[Jul-2020]: Our paper “ReARM: A Reconfigurable Approximate Reconfigurable Multiplier for Energy-Efficient Approximate Computing” has been accepted at VDAT 2020.

[May-2020]: Our paper “ACA-CSU: An Approximate Carry-Aware Compressor Selection Unit for Energy-Efficient Approximate Multipliers” has been accepted at ISVLSI 2020.

[May-2020]: Started my Summer Internship at TU Wien, working with Prof. Muhammad Shafique.

[May-2020]: Started my Summer Internship at TCS Research.

[Dec-2019]: Won a bronze medal at the Inter IIT Tech Meet 2019, held at IIT Roorkee.

[May-2019]: Our team reached the semifinals of the IICDC 2019 competition, organized by Texas Instruments and DST, India.

[May-2019]: Started my Summer Internship at IIT Ropar, working with Dr. Neeraj Goel, My research journey started here.

[Aug-2017]: Started my undergraduate studies at IIT Jodhpur.