← All organizations

AI agent development and optimization tools

Superagentic AI

Superagentic AI builds developer tools for teams designing, running and optimizing AI agents. Its SuperQode framework lets developers configure the software surrounding a coding model: tools, memory, repository access, approvals and verification. Teams can version these configurations with their code, evaluate changes and run workflows with local models or hosted providers. SuperOptiX provides optimization across agent frameworks, while SuperRadar helps developers track agent tooling. The company also offers enterprise implementations and services covering agent architecture, evaluation and optimization.

Founded and led by Shashi Jagtap, a former Apple employee, the company’s UK entity was incorporated in 2025. Its SuperOpt research explores environment-level improvements for autonomous agents rather than model retraining.

The company described itself as a company of one in its 2025 retrospective. By August 2026, it reported 97 entries across six categories in SuperQode’s Harness Hub, spanning native harnesses, coding agents, presets and ecosystem-watch records. The catalog distinguishes runnable entries from those requiring setup or awaiting integration.

super-agentic.ai

1 talk

Newest first

1 speaker at AIE

Affiliations reflect their AIE appearances, not necessarily current employment.

Messages from the stage

Inspectable execution with bounded observations

The RLM Code research playground demonstrates a workflow that returns bounded observations and controls execution budgets. Jagtap also compares it with author-maintained RLM and DSPy.RLM implementations.

Affiliations reflect each recorded session, not necessarily current employment.

Company sources · checked 2026-08-28