Jeet Sheth

Full-stack engineer with hands-on experience building production web applications, cloud-native services, and AI-powered automation systems used in manufacturing environments. Delivered systems that reduced operational analysis time 35%, cut manual process review 40%, and improved throughput via real-time prioritization pipelines. Strong in TypeScript/JavaScript, Python, SQL/NoSQL, REST APIs, AWS, and modern AI workflows (RAG, evaluation loops, embeddings).

Software Engineer at Machinecraft (2025-06 – Present)

Manufacturing AI + Automation

• Built and deployed a retrieval-augmented AI agent (RAG) that integrated production logs, SOPs, and quality datasets, reducing root-cause investigation time by 35% and improving downtime escalation speed.
• Engineered ingestion pipelines for structured + unstructured manufacturing artifacts, enabling fast semantic search and decision support across production planning and inventory workflows.
• Designed automated evaluation loops to validate AI outputs against historical production cases, reducing manual process review effort by 40% and improving consistency in throughput/OEE decisions.
• Collaborated with operations stakeholders to translate manufacturing pain points into system requirements, ensuring measurable business impact and reliability in production environments.

Software Engineering Intern at Machinecraft (2022-05 – 2022-07)

Full Time + Real Time Systems

• Developed a production system for real-time prioritization of 16,000+ inventory scans, leveraging 5 years of historical data to improve operational decision-making.
• Built a full-stack UI dashboard using React + TypeScript to visualize live inventory counts and scan details for internal manufacturing teams.
• Implemented exception-handling workflows and edge-case logic to ensure robust behavior in production systems and prevent downstream failures.

Software Engineering Intern at Vishwa Technologies Pvt. Ltd (2021-05 – 2021-07)

• Supported migration of a Transport Management System from on-prem infrastructure to AWS cloud services.
• Designed and implemented DynamoDB schema (3-table model) for high-throughput, highly available non-relational storage; configured read/write capacity scaling and CloudWatch monitoring.
• Built event-driven processing using AWS SQS, implementing batch polling to reduce ingestion costs by 15%.
• Added DLQ + redrive strategy to improve reliability and reduce downtime from 2 hours to 10 minutes, increasing processing throughput by 18%.
• Participated in code reviews, configured IAM roles and security policies and CloudWatch Alarms.

Master of Science in Computer Science – University of Texas Arlington (2023-08 – 2025-05)

Bachelor of Technology in Information Technology – Netaji Subhas University of Technology (2019-08 – 2023-05)