Available to hire
Hi, I’m Sowmith Sripadi, an AI Engineer with a strong background in Data Engineering and 3.5+ years of experience designing and deploying production-grade AI systems and enterprise data solutions.
I specialize in multi-agent architectures with LangGraph and LangChain, Retrieval-Augmented Generation pipelines using Pinecone, FAISS, and Chroma, and building reliable guardrails, telemetry, and checkpointing for scalable, safe conversational AI. I love optimizing latency, boosting model accuracy, and collaborating across product, security, and legal teams to deliver safe-by-default AI features.
Skills
Work Experience
Machine Learning Engineer at Comcast, USA
February 1, 2025 - PresentProduced LLM-powered personalization models (GPT-4, LLaMA, Amazon Titan) with LangChain RAG pipelines, driving a 15% increase in customer engagement and measurable retention improvements. Contributed to ML pipelines scaling to over 10 million daily interactions on AWS SageMaker with Kubernetes, utilizing MLflow for experiment tracking and reproducibility. Optimized inference using ONNX Runtime, quantization, and distillation, reducing p95 latency by approximately 40% (from ~220ms to ~130ms) enabling higher throughput for real-time recommendations. Integrated vector search (Pinecone, FAISS) into recommendation workflows, enabling sub-second semantic search across large-scale datasets. Collaborated closely with product and engineering teams to ensure ML models addressed key business objectives and enhanced customer experience and retention.
AI-ML Research Assistant at Florida International University
December 31, 2024 - September 5, 2025Conducted research on efficient LLM fine-tuning (LoRA, QLoRA, PEFT), reducing training cost by 35% while maintaining high accuracy in domain adaptation tasks. Developed multi-modal ML pipelines (text + vision) improving classification accuracy by 15% compared to existing baselines. Contributed to research publications on transfer learning and low-resource adaptation supporting the lab’s academic initiatives. Designed synthetic text augmentation pipelines expanding training datasets from 100k+ to 140k+ samples, improving model robustness and generalization on low-resource tasks. Mentored 5+ graduate students on best practices for reproducible ML workflows, introducing MLflow tracking that streamlined experiment comparison and improved research efficiency.
Associate Machine Learning Engineer at Accenture, India
December 31, 2022 - September 5, 2025Designed and deployed predictive analytics models for retail and finance clients, improving forecasting accuracy by 30% and reducing decision-making delays by 20%. Created time-series forecasting and anomaly detection systems using Python, SQL, and Spark enabling early anomaly detection and improved operational reliability. Delivered ML solutions with Docker, Kubernetes, and Airflow processing TB-scale datasets on AWS (S3, SageMaker) and GCP (BigQuery, Vertex AI) across multiple projects. Implemented model monitoring, SHAP/LIME explainability, and drift detection aligning deployed ML systems with internal AI fairness and governance guidelines. Developed and deployed web applications integrating ML models into production-grade systems using React (frontend), Node.js and Express (backend), improving usability and adoption by 40% among end-users.
Associate AI Engineer at Comcast
February 1, 2025 - PresentDesigned and shipped production-grade multi-agent systems using LangGraph and LangChain to power conversational workflows; achieved 15% lift in CTR, reduced no-answer rate by 11–13%, and maintained p95 latency of 130 ms. Engineered modular Python SDKs and REST services for managing prompt templates, tool registries, and agent configurations; implemented SQLite checkpointing for resumable conversations and replay tooling; developed guardrails (content filtering, function allowlists) and evaluation harnesses; deployed Dockerized services via Terraform-managed infrastructure with OpenTelemetry instrumentation.
Graduate Research Assistant at Florida International University
December 1, 2024 - October 15, 2025Collaborated with a non-profit organization on a federally funded research initiative, building Microsoft Fabric pipelines to process 22,000+ participant records and enable standardized equity analytics. Applied PySpark transformations in Databricks for schema validation, null handling, and deduplication, improving data readiness and completeness scores by 75%. Designed and optimized Power BI dashboards using star schema modeling and DAX, reducing dashboard refresh time by 50% and enhancing data usability. Deployed secure, role-based dashboards with export restrictions and dynamic filtering, increasing user engagement and research output visibility.
Data Engineer at Accenture, India
December 1, 2023 - October 15, 2025Built ADF pipelines aggregating HR data from 25+ regions, reducing reporting turnaround time from three days to under six hours with scheduled refresh and retry mechanisms. Tuned PySpark jobs in Databricks by pruning partitions and caching, reducing notebook runtimes by 67% and enabling faster transformation of employee records and policy data. Optimized Azure SQL queries using filtered indexes and execution plan tuning, decreasing dashboard response time from 20+ seconds to under 7 seconds. Integrated Power BI dataset refresh with ADF, automating daily incremental loads and improving report accuracy for compliance dashboards. Applied data masking and AES-256 encryption on PII fields to align with GDPR and HIPAA policies. Led end-to-end project delivery for the Reporting System, achieving 99% production accuracy with minimal post-launch issues.
Education
Master of Science in Computer Science at Florida International University
January 1, 2023 - December 31, 2024Master of Science, Computer Science at Florida International University
January 11, 2030 - December 1, 2024Qualifications
Industry Experience
Software & Internet, Financial Services, Retail, Education, Professional Services, Media & Entertainment, Other
Skills
Hire a AI Engineer
We have the best ai engineer experts on Twine. Hire a ai engineer in Cleveland today.