Ujjwal Dave

Software Development Manager with 16+ years of experience, including over 5 years leading high-performing teams in delivering innovative software solutions. Currently overseeing engineering operations at Khoros R&D India Pvt Ltd, focusing on diverse tech stacks and timely project completion. Expertise in strategic planning, project management, and operational efficiency, resulting in increased team productivity and successful software launches. Notable for implementing advanced technologies that enhance efficiency and foster collaboration.

• Led on-time delivery of critical features for a complex Marketing Suite SaaS platform, ensuring 99.9% uptime.
• Successfully managed multiple scrum teams and delivered 100% of the planned roadmap items QoQ along with tech debts, customer bugs and scheduled Ops maintenance and Infosec fixes.
• Multiple Manager Choice awards at IBM for individual contributions and team outcomes Best Project Award during the "PRISM" training program at Aricent

Some key projects that I worked on at Khoros.  

Building a next-generation publishing platform on top of the existing care platform, thus  providing both capabilities under a common platform.  

• Engineered a comprehensive publishing system on top of the Care platform.
• The project had a unanimous goal of reducing the customer pain point of maintaining and managing governance across two different platforms.
• Steered the integration project, combining publishing with care capabilities onto a single platform.
• Developed features from the ground up, and encountered substantial learning challenges.
• Segmented work into manageable milestones, enhancing team productivity.
• Maintained continuous communication with stakeholders using a real-time project status dashboard.
• Consistently met deadlines, boosting stakeholder confidence in team capabilities.

Link in bio, key design, and implementation highlights.  

• Developed a versatile Link-in-Bio feature for Instagram on the Khoros platform.
• Implemented a scalable micro-frontend UI for seamless integration into various applications.
• Introduced a backend Link in Bio service with an independent database for efficient updates.
• Managed content transfers to AWS S3, serving as the central datastore for images and metadata.
• Utilized a CDN to ensure global accessibility, with low-latency responses.
• Designed for adaptability, allowing dynamic changes without impacting the main application.
• Enables easy portability of the Link In Bio feature across diverse platforms.

Identifying themes in conversation for tweets, key design, and implementation  Highlights: AI/ML project. 

• Identifying themes of conversation in Twitter data, and showcasing the major themes in the data being searched by our users.
• The key challenge was to attach a theme to each and every tweet that is ingested into Khoros, without increasing the latency in a highly performant ingestion pipeline.
• We decided to develop the ML pipeline based on the Clustering (k-means) algorithm and Yake for getting theme names in Scala, and integrate that with the service that enriches the incoming tweets.
• Build the ML model using ML pipelines, using Spark ML and Spark NLP libraries, and train the ML model with new data at regular intervals to keep the model up to date. Integrate the ML model into a service that processes incoming tweets, and attach themes to them.
• The design and approach we took to implement this feature were a huge success; customers really appreciated the availability of themes of conversation for their searches, and all this happened in real time, without any delays.