.png&w=256&q=75)
Anish Jaiswal@jslanish0162317
2
Events attended
1
Submissions made
United States
1 year of experience
About me
I'm a Computer Science senior at the University of North Texas (minor in Physics) and an Undergraduate Research Assistant in the Responsible AI Lab, where I work on making AI systems more reliable and efficient in the real world. Right now I'm designing an NVIDIA-sponsored study on how sustained thermal load affects vision-language model inference on the NVIDIA DGX Spark, and developing an open-source Python benchmarking tool to measure latency and GPU performance under load. My path into AI started with hardware and robotics. I began as a computer repair technician and worked as a Project Designer and Robotics Trainer at the Robotics Association of Nepal before moving into software and AI in the U.S. That background still shapes how I build — I'm drawn to systems that connect code to the physical world. I've built an AI-powered interview-coaching app (HireMate) using large language model APIs, an agriculture robot with IoT soil sensors, and a range of full-stack and database projects. I work primarily in Python, C++, and JavaScript. My interests center on trustworthy AI, LLM and vision-language systems, robotics, and quantum machine learning. Across all of it, I'm driven by the same goal: using technology to build dependable, meaningful solutions that create real impact for people and communities.
🤝 Top Collaborators
🤓 Latest Submissions
SimToReal Agent
Problem: Re-tasking a factory robot is slow and risky. AI can write a robot program in seconds, but a program that looks valid can still drop a part, hit the table or reach for something out of range. On a real line that means scrap, damaged tooling and downtime, so nobody should run an untested AI program on a real machine. User: the cell operator or automation engineer who must sign off that a new robot program is safe. How it works: The operator types a task such as "Sort both parts by color". An LLM on AMD Token Factory turns it into explicit goals and writes eight candidate pick-and-place programs, each checked against the allowed actions and objects. All eight run in parallel in a MuJoCo simulation of a Franka Panda arm, which records video, a final camera image, collisions and dropped parts. A safety-first scorer rejects any program that collides, drops a part or misses the goal, and ranks the rest. A vision model on AMD checks the best program's final image; if it disagrees with the simulator or cannot tell, the plan is flagged for human review. The operator reviews the evidence and approves, and only then is the program exported. The app never commands a real robot. AMD: All inference runs on AMD Token Factory: DeepSeek-V4-Flash plans, DeepSeek-V4.1-Flash reviews images. An AMD panel shows the models and measured times for every run. Results: In live runs, 24 AI-written programs for three tasks were tested. 12 passed and 12 unsafe or impossible ones were caught in simulation, each with a readable reason. For an out-of-reach part, all eight were rejected and no plan was offered. Planning eight programs took 6 to 8 seconds.
18 Oct 2026
