AI Tools Engineer (Autonomous Agents / Devin-like Systems)

Job Title: AI Tools Engineer (Autonomous Agents / Devin-like Systems)

Role Overview

We are looking for an AI Tools Engineer to design and build autonomous coding agents similar to Devin AI. This role focuses on developing intelligent tools that can plan, write, debug, and deploy code with minimal human intervention, enhancing developer productivity through AI-driven systems.

============================================

Key Responsibilities

Optimize AI-powered tools for end-to-end software development workflows

Integrate LLMs (such as GPT-4, Claude 3.5, or Llama 3) into developer tools and platforms

Design and implement secure, sandboxed execution environments (e.g., Docker/WASm) for agent-led code execution

Design multi-step reasoning, planning, and tool-usage capabilities for autonomous agents

Develop intelligent systems for code generation, testing, and automated self-healing/debugging

Work with APIs, databases, and deployment pipelines to enable seamless automation

Improve model performance through fine-tuning, prompt engineering, and Reinforcement Learning from Compiler Feedback (RLCF)

Collaborate with product and engineering teams to deliver AI-driven developer tools

===========================================

Required Skills

Strong programming skills in Python / JavaScript / TypeScript

Experience with machine learning / deep learning frameworks (PyTorch, TensorFlow)

Solid understanding of LLMs, AI agents, and Chain-of-Thought (CoT) / ReAct prompting techniques

Familiarity with software engineering best practices (Git, CI/CD, testing)

Knowledge of APIs, microservices, and cloud platforms (AWS/GCP/Azure)

Understanding of Vector Databases (Pinecone, Milvus, or Weaviate) for long-term agent memory

============================================

Preferred Qualifications

Experience with AI agent frameworks (LangChain, LangGraph, CrewAI, AutoGen, etc.)

Background in reinforcement learning or AI planning systems

Experience building AI-powered developer tools or IDE integrations (VS Code Extensions, JetBrains plugins)

Exposure to distributed systems

Experience with Abstract Syntax Trees (AST) or static code analysis for better agent understanding of codebases

============================================

Nice-to-Have

Contributions to open-source AI projects

Experience with AI coding assistants (e.g., GitHub Copilot, Cursor)

Research experience in AI or NLP