Kaushik Golithdka
Redefining AI @ Autodesk | Applied AI R&D | Full-Stack AI Engineer | Agentic AI, LLM Fine-Tuning, RAG | MS CS & AI (USC)
- Role
- Software Engineer 2 - Generative Ai at Autodesk
- Location
- San Francisco, CA, US
- LinkedIn followers
- 500 followers
About Kaushik Golithdka
Innovative Full-Stack AI Engineer with 6.5+ years of experience (4.5+ years full-time, 2+ years internships) driving groundbreaking AI solutions that redefine developer productivity. Specializing in Agentic AI systems, LLM fine-tuning, multi-agent orchestration, and retrieval-augmented generation (RAG), I design and implement AI systems that compress complex multi-step workflows from hours to minutes, automate repetitive tasks, and create measurable business impact at scale.I thrive at the intersection of AI research and applied engineering, rapidly transforming experimental concepts like agentic subagents, developer debugging agents, and AI-driven code orchestration into production-ready solutions. Reporting directly to the Head of Architect, VP, and CTO, my work has immediate visibility and strategic impact on Autodesk’s AI vision.Core Competencies:* Agentic AI & Multi-Agent Orchestration (main agent → subagents → MCPs)* Full-Stack AI Engineering: Backend orchestration + Frontend chatbot systems* LLM Fine-Tuning & Domain Adaptation* Retrieval-Augmented Generation (RAG) with embedding optimization* AI Workflow Automation & Developer Productivity Tools* Scalable Backend & Microservices (Python, Spring Boot, Docker, Azure)Highlights & Impact:* Built MCP orchestrator with main/subagent architecture, enabling parallel execution of tasks and reducing complex workflows by ~70–100x, dramatically improving developer productivity.* Developed subagents that autonomously code, test, validate, and review features 24x7, cutting development cycles by ~50–80%.* Created a Developer Debugging Agent that monitors coding, staging, and production deployments, reducing runtime errors and preventing system failures by ~60–80%.* Built AI-powered PR review agents that analyze full context and provide high-accuracy recommendations, accelerating code review turnaround by ~5–10x.* Applied latest LLM techniques to optimize AI decision-making and task automation pipelines, improving overall AI reliability and efficiency.* Delivered high-visibility AI features at 1–2 week cycles, ensuring rapid prototyping, iteration, and deployment across Autodesk.I’m passionate about building AI systems that empower developers, accelerate decision-making, and create measurable business impact, combining research-grade innovation with production-ready engineering. Contact: g••••••••@gmail.com
Experience
Software Engineer 2 - Generative Ai
May 2024 — Present · San Francisco, CA, US
Driving high-impact AI solutions that redefine developer productivity. As part of the Applied AI R&D team, I rapidly research, prototype, and deploy full-stack agentic AI systems, working directly with the Head of Architect, VP, and CTO to influence Autodesk’s AI strategy.Key Contributions:* Agentic AI Orchestration: Built an MCP gateway with main/subagent hierarchy, executing tasks in parallel and reducing multi-hour workflows to minutes (~70–100x improvement).* Autonomous Subagents: Developed agents that write, test, validate, and review code continuously (24x7), accelerating development cycles by ~50–80%.* Developer Debugging Agent: Created a background agent that tracks developer activity from coding → staging → production, preventing runtime errors and reducing deployment issues by ~60–80%.* AI-Powered Code Review: Implemented LLM-based PR review system providing full-context feedback, improving review turnaround and code quality by 5–10x.* Rapid Prototyping: Delivered high-visibility features at 1–2 week cadence, leveraging the latest Applied AI research in production.* LLM Fine-Tuning & RAG: Optimized domain-specific embeddings and decision-making workflows, improving AI precision and reliability.* Strategic Leadership Impact: Solutions developed are directly visible to VP/CTO, shaping Autodesk’s AI productivity initiatives and future roadmap.Business Impact:* Reduced complex engineering workflows by ~70–100x* Saved thousands of engineering hours annually* Increased developer productivity and confidence in deployments* Accelerated feature development cycles by 50–80%
Education
Kendriya Vidyalaya
High School, Computer Science
2006 — 2015
University of Southern California
Master of Science - MS, Computer Science and AI
Bangalore University
Bachelor’s Degree, Computer Science And Engineering
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.