Lucas

As a Claude Ambassador working with Anthropic, I help represent the company's products and its flagship AI model, Claude, on the Indiana University campus.

My role focuses on promoting responsible AI development by engaging peers with workshops, hackathons, and collaborative events. I also create content to share my experiences with Claude and partner with campus organizations to expand awareness of AI’s applications in problem-solving and learning.

Within this program, I serve as the president of the Claude Builder Club, an official student organization sponsored by Anthropic, and lead its executive board. Some highlights include 7+ workshops hosted by us and featuring team members from Anthropic, 2 hackathons with 200+ students and $1.5k in prizes, and more - hint: take a look at the related projects :)

I provide independent software engineering and technology consulting services to early-stage startups and organizations. I am actively taking clients; if you are interested, feel free to get in touch!

• Consult directly with founders and teams to scope requirements, translating product requirements documents (PRDs) and design systems into scalable system architecture and technology recommendations.
• Build full-stack products end-to-end, including React Native mobile apps and backend services with LLM integration.
• Design and maintain cloud infrastructure and databases (PostgreSQL, AWS, Supabase), applying security-first principles for handling sensitive user data.
• Support go-to-market efforts where needed, including marketing strategy and partnership outreach.

Selected clients

• Percipient
• Rally
• Dumb Politics
• PlanItUni
• Computational Thinking Lab (CTL)
• Packt
• Google
• Anthropic

As Co-Founder & CTO across both ventures, I directed end-to-end product architecture, engineering, and AI system design. My entrepreneurial journey began with Hubble (later rebranded as Bobby) in AI-driven candidate recruitment. Navigating early beta testing and customer discovery revealed critical product–market fit lessons, leading to an intentional pivot into legal technology to co-found Goliath.

Hubble (later Bobby) — AI-Driven Candidate Profiling & Matching

• Co-founded and served as CTO alongside Sehoon Park, building an AI-powered hiring platform designed to connect students with early-stage startups through voice-based profiling and intelligent matching.
• Architected AI voice-agent workflows and matching pipelines to capture candidate experience, communication style, and values beyond traditional static resumes.
• Grew the platform to 300+ student sign-ups and onboarded early-stage startups during initial beta testing to validate demand in the student–startup hiring space.
• Accepted into the IU Innovates startup incubator, iterating on product features and market strategy through mentorship and user feedback.
• Competed in and won collegiate pitch competitions, including Start Tank and InnoQuest.
• The Strategic Pivot: After extensive user testing and feedback analysis, we identified key product–market fit and distribution hurdles in student recruitment. Rather than forcing a misaligned product, we made the strategic decision to sunset Bobby (see meetbobby.vercel.app) and redeploy our technical expertise in agentic AI and workflow automation toward a high-trust, high-value domain: legal tech.
• Explore the venture: Hubble (later Bobby) Project Details

Goliath — Multi-Agent AI Legal Automation

• Co-founded Goliath as CTO, owning full-stack architecture, agent pipelines, and system security for an AI workflow platform tailored for estate planning law firms.
• Designed secure multi-agent architectures and hybrid retrieval (RAG) systems that record client meetings, synthesize structured notes, and generate first-pass legal drafts with attorney-in-the-loop oversight.
• Engineered rigorous guardrails, deterministic criteria resolution, and security models required for high-trust legal environments.
• Competed in and won the Elmore Startup Clinic pitch competition (IU Maurer School of Law).
• Currently running paid pilots with law firms across Indiana, collaborating closely with practicing attorneys while expanding capabilities into automated billing and sentiment analysis.
• Explore the venture: Goliath.ai Project Details and Goliath's Website

• In Fall 2024, investigated quantum convolutional neural networks (QCNNs) within hybrid classical-quantum models. Applied Python (especially TensorFlow Quantum) and quantum computing concepts to study QCNNs

• In Fall 2025, Conducted academic research on the application of Large Language Models to clinical data (specifically EHR data), information extraction, and diagnosis prediction. Was the lead author on a paper associated with such research, which was submitted to the International Joint Conferences on Artificial Intelligence (IJCAI). Also operated a SLURM-based computing cluster to enable distributed training and inference using DeepSpeed and tensor parallelism, while implementing both learned and traditional NLP evaluation metrics. .

• Presented findings to both technical stakeholders and non-technical audiences in a poster session.

• Currently, I'm an AI/ML researcher for the Computational Thinking Lab

• Founded Entropic, an organization that empowers individuals and organizations in the domains of technology and entrepreneurship.
• Established a branch of Entropic at Ivy Tech (Entropic@IT) and expanded to Indiana University (Entropic@IU), as student-led organizations that host events, networking opportunities, and technical workshops.
• Facilitated and mentored collaborative projects, including full-stack web apps, management systems for local businesses, and software-focused initiatives led by student teams.
• Created technical and entrepreneurship study guides, eBooks, and curated roadmaps for students to build practical skills with free, high-quality resources.
• Grew an engaged online community through a structured Discord server and social media channels (Instagram, TikTok, LinkedIn), along with social media content production.
• Co-hosted 2 Hackathons, each with 200+ participants and 1.5k+ in prizes

Worked on AI systems for healthcare Utilization Management (UM) at Elevance Health, focusing on clinical criteria automation, predictive modeling, and production ML infrastructure.

Some of the stuff I did:

• Improved a claims denial risk scoring system through feature engineering, raising PR-AUC from 0.82 to 0.90, and implemented model routing to reduce inference latency by 10%.
• Integrated Fiddler across 15+ production machine learning models (primarily XGBoost and ensemble classifiers) for case-detail probability predictions to monitor data drift and model performance. Analyzed telemetry across 5 production services to identify opportunities for latency reduction, precision gains, and infrastructure cost savings.
• Contributed to an AI-assisted Utilization Management automation platform:
  • Built a configurable DAG-based workflow engine that coordinates hybrid retrieval (semantic and BM25 RAG), structured document extraction using OCR and LLMs, and deterministic clinical criteria resolution.
  • Developed a multimodal, long-context AI agent that analyzes unstructured clinical policy documents and converts them into executable YAML DAG workflows.
• Deployed and orchestrated production AI/ML microservices on Kubernetes using Kubeflow, supporting containerized model serving across a team of 13+ data scientists and engineers.

Worked as an Undergraduate Instructor for the class C343 - Data Structures & Algorithms at Indiana University. My responsibilities included teaching labs, grading assignments, designing test cases, and supporting students.

• Worked on infrastructure automation, internal tooling, and modernization efforts supporting enterprise university systems and web services. Automated server administration workflows using Bash and Python, including MySQL backups, monitoring tasks, and CI/CD improvements.
• Maintained Linux-based enterprise environments running Apache, MySQL, and PHP, with a focus on containerization using Docker and Kubernetes.
• Developed an internal dependency management platform in Java using Spring Boot REST APIs and a React-based administrative interface.
• Implemented monitoring and observability solutions with Grafana to support system reliability and operational visibility.
• Built an MCP server using Spring AI, enabling secure OAuth2-authenticated interaction with internal APIs through LLM platforms such as ChatGPT and Claude.

• Collaborated with a team of eight engineers to develop software solutions for TrueLeap, a non-profit focused on accessible e-learning.
• Used Python, PyTorch, Pinecone, and LangChain to design a chatbot powered by LLMs and Retrieval-Augmented Generation.
• Worked on goals involving driving hallucinations below 3% and improving response accuracy to ~95% using FastAPI services and better context retrieval.

• Worked on retail-focused AI/ML and data engineering initiatives supporting e-commerce operations, inventory management, and analytics workflows. Managed and optimized AWS-hosted infrastructure and data systems, including a retail inventory database containing more than 600 SKUs.
• Built and maintained scalable ETL pipelines using Python, Apache Airflow, AWS Kinesis, Pandas, and MySQL (AWS RDS) to process inventory and sales data for downstream analytics and machine learning applications.
• Performed data exploration, reporting, and visualization using Power BI, Excel, and AWS SageMaker to support operational insights and model development.
• Developed AI-driven systems including retrieval-augmented generation (RAG) applications and a real-time computer vision classification model for physical product recognition and retail automation workflows.