Rama Chandra@ramachandraai

AI Evaluation Lab is an independent engineering project focused on making AI agents easier to test, verify and improve. For the AMD AI League, I am developing a shared foundation for source-grounded information retrieval and test-verified code repair. The retrieval workflow is designed to extract specific answers from documents or websites, preserve the supporting sources, and return no answer when the available evidence is insufficient. The code-repair workflow will reproduce a failure, propose a focused patch, and check it against relevant tests before reporting the outcome. The planned implementation uses Python, open-source models and AMD ROCm, with Docker containers tailored to each match's requirements. Evaluation will cover answer correctness, citation accuracy, repair success, regression failures, latency and GPU memory use. Separate test cases will help check whether improvements generalize beyond the public examples. This solo entry builds on my work in AI evaluation, benchmark development and OmniEdge RouteIQ from AMD ACT II. The project is currently in development; AMD runtime validation and official challenge scores are pending.

FailureTwin AI is an AI-powered pre-mortem decision workspace designed to help founders, product managers and operations teams discover how a plan could fail before committing time, money or resources. Instead of simply generating a plan, FailureTwin challenges it from four perspectives: Customer, Operations, Finance and Risk. Users describe a launch, workflow, automation or business decision, answer focused follow-up questions, and receive a structured analysis of weak assumptions, evidence gaps and high-impact risks. FailureTwin then connects those risks into failure chains, showing how one weak assumption can cascade into operational, financial or customer consequences. The Decision Room recommends safeguards that users can accept or reject while the readiness score updates from an immutable baseline, preventing scoring drift. The workflow ends with an explainable Go, Revise or Stop Decision Brief containing the most important risks, safeguards, confidence level and next actions. The project includes guided scenarios for AI customer support, delivery expansion and hospital claim automation, plus custom simulations. A deterministic fallback engine keeps the experience reliable if the live AI path is unavailable. Our goal is simple: help teams find expensive mistakes while they are still inexpensive to fix.

OmniEdge RouteIQ is a general-purpose AI agent built for AMD ACT II Track 1. It processes eight benchmark categories: factual knowledge, mathematical reasoning, sentiment classification, summarization, named entity recognition, code debugging, logical reasoning, and code generation. The scoring agent runs inside a public Linux AMD64 Docker container. At runtime, it reads tasks from /input/tasks.json, uses only the models supplied through ALLOWED_MODELS, routes all requests through the evaluator-provided FIREWORKS_BASE_URL, and writes structured answers to /output/results.json while preserving each task_id. RouteIQ follows a Gemma-first model policy. When an approved Gemma 4 model is available, it is selected before permitted fallback models. The implementation avoids hardcoded credentials and model assumptions, supports evaluator-controlled API access, and writes progressive results so completed work survives interruptions. The wider project also includes a FastAPI routing backend and a Next.js command dashboard for demonstrating route decisions, provider status, simulations, metrics, and audit visibility. The benchmark container remains self-contained so automated judging can pull and run it independently.

Six Matches. One leaderboard. Build on AMD, earn points and compete for prizes.

💻 Online Hackathon | ⏳ Build: 48 hours | 📅 September 25–27, 2026 The official IBM Bob 2.0 hackathon. Build what's next in AI-assisted development, alongside the developers and experts shaping it. ⭐ Get help from experts throughout the event 🤝 Join a team or build solo 💰 Prize pool: $12,000 🧑🏻‍💻 Register before the Kick-Off Stream to start building with Bob

🌎 Online Hackathon | 📅 August 3–10, 2026| 💻 Powered by native.builder The official native.builder hackathon, hosted by NativelyAI and lablab.ai. Join founders, developers, designers, and builders to create real AI-native applications powered by native.builder: from idea to deployed product, fully in the cloud. 💻 This hackathon is fully online. No travel required.

To participate in the hackathon, simply click Sign up with AMD. If you are not already a member of the AMD AI Developer Program (ADP), you will need to create an account. ADP members can access AI tutorials, experts, and community support. In addition, new members can claim a free month of DeepLearning.ai pro, $100 in AMD GPU cloud credits, and $50 in Fireworks AI API credits. All hackathon participants will receive on hackathon launch additional compute and API credits for their projects. Ready to build what's next on AMD? Sign up with AMD below to secure your spot.