• Build the end-to-end pipeline: recording → transcription → summarization → coaching report • Auto-generate Pre-Diagnosis Reports from survey responses and startup data • Design expert matching algorithms using startup profiles and diagnosis results • Architect and implement Multi-Agent systems • Fine-tune LLMs and train domain-specific models • Design prompts, build evaluation frameworks, manage model versions, and run A/B tests

• Architect & ship: Design end-to-end Python services (batch + streaming) that ingest model outputs, run constrained optimization, and surface real-time dispatch decisions. • Model & simulate: Build/extend ML models (gradient-boosting, deep learning, OR-Tools) and run time-horizon simulations to quantify cost vs. service-level trade-offs. • Operationalize: Automate training, validation, A/B rollout, and monitoring (SageMaker / Airflow). • Lead & collaborate: Partner with Product, Ops, and Data Engineering; mentor a small squad of ML engineers; present findings to execs. • Continuously improve: Instrument NPS / cost telemetry, identify failure modes, and iterate.

• Work across diverse GenAI platforms like AWS, Salesforce, Oracle, Snowflake, MS Copilot, and other 3rd party GenAI platforms and libraries. • Automate workflows involving extraction of complex, multimodal unstructured content from variety of sources in to highly accurate and reliable structured content using platforms like AWS Textract and Bedrock • Design and build MCP hosts, clients and servers • Establish and use frameworks for automated LLM testing • Create regression test suites to detect drift or prompt breakage • Integrate with internal and external web services using secure authentication and authorization mechanisms • Adopt and ensure safe practices to protect against prompt injections, jailbreaks, and conform to enterprise security guidelines • Design, develop, and deploy production-grade traditional ML models (e.g., regression, classification, clustering, recommender systems) for a variety of business use cases. • Design, maintain, and optimize end-to-end AI/ML pipelines including data ingestion, training, evaluation, deployment, and monitoring on cloud infrastructure (e.g., AWS or equivalent) • Ensure AI/ML solutions are scalable, reliable, secure, and cost-effective within cloud environments • Create reusable components, frameworks, and best practices to accelerate AI development • Partner with data scientists, architects, product managers, business stakeholders and technical teams across organization to align AI solutions with organizational goals. • Provide hands-on technical support and mentorship to technical teams across the enterprise.

• Define the companys ML strategy: where ML should be applied across products, what infrastructure is required, and how to approach build vs. buy decisions. • Design and build production ML systems end-to-end — including data pipelines, model training workflows, evaluation frameworks, and inference serving. • Establish rigorous evaluation methodology to measure model quality, detect regressions, and support data-driven iteration. • Own the data strategy: determine what data is needed, how it should be labeled, how feedback loops are structured, and how models continuously improve. • Partner closely with product and backend engineers to integrate ML into customer-facing systems. • Write production-quality code within the existing codebase and contribute to architectural decisions. • Over time, help recruit, mentor, and lead the ML team as the function expands.