SiMa Launches Agentic Development Environment for Physical AI

2026-06-16 · Source: Big Data & AI News - EE Times · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Software Development & Engineering · Depth: Intermediate, short

Summary

SiMa.ai, an edge AI chip company, has launched Palette Neat, an agentic development environment within its Palette SDK, designed for physical AI system developers using the Modalix MLSoC chip. This new environment combines an execution library with an agentic workflow to significantly reduce development time for applications like robots, drones, healthcare, and industrial automation. SiMa CEO Krishna Rangasayee stated that developers can design systems in plain English, condensing typical 6-12 week coding efforts into days or hours. The platform addresses challenges in embedded AI, including sensor integration and data parsing, by allowing agents to design SiMa-based AI systems based on developer prompts regarding sensors, accuracy, and latency. Palette Neat also features code conversion capabilities, enabling agents to analyze and optimize existing Nvidia Cuda kernels for SiMa hardware, bridging Cuda's perceived moat. SiMa shipped approximately 1,000 units of its production-qualified SoM last year, and 20 customers have already engaged with Palette Neat.

Key takeaway

For AI Engineers and Machine Learning Engineers developing physical AI systems on edge devices, SiMa.ai's Palette Neat offers a significant acceleration in your development cycle. If you are struggling with sensor integration, data parsing, or optimizing code for embedded hardware, this agentic environment can condense weeks of effort into days. You should explore using plain English prompts to design systems or utilize its code conversion capabilities to adapt existing Nvidia Cuda kernels for SiMa hardware, potentially reducing time-to-production and friction in hardware transitions.

Key insights

SiMa.ai's agentic development environment accelerates physical AI system design and code optimization for edge devices.

Principles

• Software quality is crucial for silicon company success.
• Abstracting AI access scales adoption in less proficient markets.
• Agentic workflows can optimize code better than humans.

Method

Developers prompt an agent with system details, sensors, and constraints; the agent then designs an optimized SiMa-based AI system or converts existing code.

In practice

• Use plain English prompts to design edge AI systems.
• Convert existing Nvidia Cuda kernels to SiMa hardware.
• Integrate SiMa's IP with Synopsys' architecture evaluation platforms.

Topics

• Edge AI
• Agentic AI
• Physical AI Systems
• SiMa.ai Modalix MLSoC
• Embedded Systems Development
• Code Optimization

Best for: AI Engineer, Machine Learning Engineer, Robotics Engineer

Related on AIssential

• NVIDIA Vera CPU Sets a New Standard for Agentic Workloads in AI Factories — The NVIDIA Vera CPU optimizes agentic AI and reinforcement learning by making CPU execution a high-performance, critical path within the AI loop.
• Meta to Spend Billions on AMD Gear, AI Scare Trade Continues — Diversification in AI chip procurement and specialized AI agents are reshaping market dynamics and software valuations.
• NVIDIA GTC Taipei 2026 Moments: Agents, AI PCs, and Physical AI — NVIDIA is driving AI innovation through new hardware, software, and agent development tools.
• Edge AI for Smart Cities — Agentic AI at the edge transforms smart cities into cognitive cities by enabling real-time sensing, reasoning, and autonomous action.
• The Nvidia AI PC, Project Solara, Microsoft AI — The AI landscape is shifting from local inference PCs to cloud-centric agent platforms and customizable enterprise models.
• When Data Stops Being Code and Starts Being Conversation (Ep. 297) — AI agents are transforming data generation from scripted pipelines to conversational negotiation, enhancing realism and accessibility.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Big Data & AI News - EE Times.