Agentic AI Engineer : CraftifAI Orbit

CraftifAI is seeking an Agentic AI Engineer to build the intelligence layer of CraftifAI Orbit, our GenAI-powered platform for automating code generation, integration, validation, and debugging for Edge AI, FPGA, embedded systems, and hardware-aware software workflows. The role requires strong hands-on capability in LLM agents, AI-assisted code generation, RAG, tool calling, reasoning systems, and evaluation frameworks applied to real engineering domains such as Edge AI pipelines, FPGA design flows, SDK integrations, build systems, and deployment automation.

• Design and build LLM-powered coding agents for code generation, code modification, refactoring, debugging, validation, and documentation across Edge AI, FPGA, and embedded workflows.

• Convert user requirements into implementation plans, source code, configuration files, test cases, build scripts, deployment artifacts, and validation reports.

• Build repository-aware agents that inspect project structures, understand dependencies, retrieve relevant files, generate patches, and validate generated outputs.

• Develop tool-calling systems for Python tools, SDKs, compilers, simulators, synthesis/P&R tools, log parsers, file operations, APIs, and internal automation services.

• Build RAG pipelines over technical documentation, reference designs, SDK manuals, hardware specifications, codebases, build logs, and engineering knowledge bases.

• Implement reasoning workflows for planning, context selection, tool selection, code validation, failure diagnosis, iterative repair, and final response generation.

• Create evaluation frameworks to measure agent accuracy, code quality, tool-call correctness, retrieval relevance, build success rate, regression safety, and hallucination reduction.

• Implement guardrails against hallucinated APIs, invalid code, unsafe tool usage, incorrect hardware assumptions, prompt injection, and unreliable reasoning.

• Build observability for agent runs, including traces, tool-call timelines, execution logs, intermediate outputs, validation reports, and failure summaries.

• Bachelor's or Master's degree in Computer Science, AI, Machine Learning, Electrical Engineering, Electronics, Embedded Systems, or a related technical field.

• Mandatory eligibility: Candidate must be from a Tier-1 engineering or computer science institution. Applications from non-Tier-1 institutions will not be considered.

• Minimum 3+ years of experience in AI/ML, software engineering, AI-assisted engineering automation, or production GenAI systems.

• Strong hands-on experience with LLMs, agentic AI systems, tool calling, RAG pipelines, and AI-assisted code generation.

• Experience with OpenAI, Claude, Gemini, Llama, Mistral, LangGraph, LangChain, AutoGen, CrewAI, LlamaIndex, OpenAI Agents SDK, or custom agent systems.

• Strong understanding of structured outputs, function calling, JSON Schema, Pydantic, API integrations, stateful workflows, and agent orchestration.

• Hands-on experience with coding agents for code generation, code review, code repair, debugging, refactoring, test generation, and documentation generation.

• Strong Python skills, Git-based workflows, repository navigation, patch generation, dependency analysis, build validation, and Linux/Ubuntu environments.

• Ability to understand complex engineering documentation, logs, APIs, SDKs, build systems, hardware/software interfaces, and system-level constraints.

• Experience in Edge AI, FPGA, embedded systems, SoC platforms, accelerator SDKs, or hardware-aware software development.

• Exposure to FPGA workflows such as HDL generation, IP integration, simulation, synthesis, place-and-route, timing constraints, vendor toolchains, or reference designs.

• Exposure to Edge AI workflows such as model optimization, quantization, runtime configuration, inference pipelines, deployment artifacts, and performance validation.

• Experience building agents for engineering automation, build-fix loops, log-driven debugging, test generation, static analysis, AST parsing, symbol search, dependency graphs, or language-server-based context.

• Familiarity with AI coding tools such as Cursor, GitHub Copilot, Aider, SWE-agent, OpenHands, Claude Code, Replit Agent, or similar systems.

• Knowledge of MCP-style tool integration, secure tool execution, agent observability, guardrails, and LLM reliability engineering.

Must Have: Agentic AI, LLMs, AI Code Generation, Coding Agents, Python, Tool Calling, RAG, Vector Databases, Context Engineering, Prompt Engineering, JSON Schema/Pydantic, Git, Linux/Ubuntu, Evaluation Frameworks, API Integration.

Strong Advantage: Edge AI, FPGA, Embedded Systems, HDL, Hardware SDKs, Build Automation, Log Analysis, Code Repair Agents, Model Optimization, Quantization, Synthesis/P&R Workflows.

The ideal candidate is an AI engineer who can build reliable coding agents for real engineering workflows, not prompt-based demos. They should design agents that understand technical requirements, retrieve engineering context, generate and modify code, call tools, analyze failures, validate outputs, and improve through evaluation. Candidates combining strong agentic AI and code-generation experience with Edge AI, FPGA, embedded systems, or hardware automation exposure will be highly relevant for CraftifAI Orbit.