open to work·SF Bay Area

Suhaas Teja

$ whoami — AI/ML Engineer · Edge Inference & LLM Systems

llm inference · quantization · multimodal ai · edge ai

I'm an AI/ML engineer who likes making machine learning fast, reliable, and genuinely useful — lately, running LLMs efficiently on edge hardware. Before that, responsible-AI evaluation and applied research, plus an M.S. in Computer Engineering from San José State.

email linkedin github x instagram

section 01

experience

Edge AI, enterprise GenAI, and applied research.

Pronto Networksedge LLM inference · 150+ brands
Jun 2026 - Present
Developing and deploying AI models on edge networking hardware, enabling on-device inference across cloud-managed 4G/5G routers, WiFi access points, and the PC64 edge AI appliance on Pronto's AIOS platform.
visit ↗
LEVCOMdeterministic edge AI · FPGA × GPU
May 2026 - Present
Engineering AI and perception systems for robotics, industrial vision, and autonomous platforms: real-time vision pipelines and inference with bounded latency on AMD/Xilinx FPGAs.
visit ↗
Humanaresponsible AI · LLM evaluation & red-teaming
May 2025 - Aug 2025
Improved observability of internal AI systems by 70% through framework-agnostic RAG evaluation workflows, synthetic data generation, and red-teaming pipelines in support of responsible AI initiatives.
visit ↗
SJSU King Libraryagentic RAG · applied AI research
Mar 2024 - Dec 2025
Launched KingbotGPT, a production agentic RAG chatbot providing after-hours reference support to 30,000+ students and faculty; owned evaluation design and reliability. Led data curation for the 2025 Landscape of Responsible AI Across California Higher Education Institutions report.
kingbotGPT landscape of AI report visit ↗
ZestIoTreal-time IoT dashboards · Boeing HorizonX award
Mar 2022 - May 2023
Developed real-time operations dashboards and asset tracking for GroundRadar, ZestIoT's airport operations platform deployed at Hyderabad, Delhi, and Bengaluru international airports, supporting 600+ daily flights and 9,000+ ground personnel.
visit ↗

Pronto Networksedge LLM inference · 150+ brands
Jun 2026 - Present
Developing and deploying AI models on edge networking hardware, enabling on-device inference across cloud-managed 4G/5G routers, WiFi access points, and the PC64 edge AI appliance on Pronto's AIOS platform.
visit ↗
LEVCOMdeterministic edge AI · FPGA × GPU
May 2026 - Present
Engineering AI and perception systems for robotics, industrial vision, and autonomous platforms: real-time vision pipelines and inference with bounded latency on AMD/Xilinx FPGAs.
visit ↗
Humanaresponsible AI · LLM evaluation & red-teaming
May 2025 - Aug 2025
Improved observability of internal AI systems by 70% through framework-agnostic RAG evaluation workflows, synthetic data generation, and red-teaming pipelines in support of responsible AI initiatives.
visit ↗
SJSU King Libraryagentic RAG · applied AI research
Mar 2024 - Dec 2025
Launched KingbotGPT, a production agentic RAG chatbot providing after-hours reference support to 30,000+ students and faculty; owned evaluation design and reliability. Led data curation for the 2025 Landscape of Responsible AI Across California Higher Education Institutions report.
kingbotGPT landscape of AI report visit ↗
ZestIoTreal-time IoT dashboards · Boeing HorizonX award
Mar 2022 - May 2023
Developed real-time operations dashboards and asset tracking for GroundRadar, ZestIoT's airport operations platform deployed at Hyderabad, Delhi, and Bengaluru international airports, supporting 600+ daily flights and 9,000+ ground personnel.
visit ↗