Indrajit S.
Building CellAssist: AI receptionist live in 54+ Indian clinics, 6,000+ patient calls answered · Co-Founder & CTO, CellStrat · Filming the road to 100 — Mission 100 Clinics
India
14+ years of work experience
Nov 20233 yrsPresent
Bangalore Urban, Karnataka, India
𝐒𝐩𝐞𝐚𝐫𝐡𝐞𝐚𝐝𝐞𝐝 𝐞𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞-𝐬𝐜𝐚𝐥𝐞 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 based 𝐀𝐈 𝐬𝐨𝐥𝐮𝐭𝐢𝐨𝐧𝐬 using 𝐬𝐭𝐚𝐭𝐞-𝐨𝐟-𝐭𝐡𝐞-𝐚𝐫𝐭 𝐋𝐚𝐫𝐠𝐞 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐌𝐨𝐝𝐞𝐥𝐬 (𝐋𝐋𝐌𝐬) for 𝐩𝐫𝐨𝐝𝐮𝐜𝐭 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭 and 𝐢𝐧𝐭𝐞𝐥𝐥𝐢𝐠𝐞𝐧𝐭 𝐬𝐞𝐫𝐯𝐢𝐜𝐞 𝐝𝐞𝐬𝐢𝐠𝐧 across 𝐦𝐮𝐥𝐭𝐢𝐩𝐥𝐞 𝐢𝐧𝐝𝐮𝐬𝐭𝐫𝐲 𝐯𝐞𝐫𝐭𝐢𝐜𝐚𝐥𝐬.
- 𝘋𝘦𝘴𝘪𝘨𝘯𝘦𝘥 𝘢𝘯𝘥 𝘪𝘮𝘱𝘭𝘦𝘮𝘦𝘯𝘵𝘦𝘥 𝘮𝘶𝘭𝘵𝘪-𝘢𝘨𝘦𝘯𝘵 𝘈𝘐 𝘧𝘳𝘢𝘮𝘦𝘸𝘰𝘳𝘬𝘴 including proprietary 𝘙𝘦𝘢𝘭 𝘌𝘴𝘵𝘢𝘵𝘦 𝘈𝘨𝘦𝘯𝘵 𝘴𝘰𝘭𝘶𝘵𝘪𝘰𝘯𝘴 using advanced orchestration technologies:
@ LangChain
@ LangGraph
@ CrewAI
@AutoGen for complex workflow automation
𝐏𝐢𝐨𝐧𝐞𝐞𝐫𝐞𝐝 𝐚𝐮𝐭𝐨𝐧𝐨𝐦𝐨𝐮𝐬 𝐚𝐠𝐞𝐧𝐭𝐢𝐜 𝐰𝐨𝐫𝐤𝐟𝐥𝐨𝐰 𝐬𝐲𝐬𝐭𝐞𝐦𝐬 that streamline complex business processes, reducing manual intervention by 45%, and enabling intelligent decision-making at scale.
𝐀𝐈 𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 & 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭
-Built robust AI service architectures using FastAPI stack for high-performance, production-ready API development with seamless scalability and enterprise-grade reliability.
- Integrated advanced search and analytics capabilities through AWS OpenSearch, enabling efficient content discovery and real-time analytics across enterprise applications.
- Developed comprehensive Model Context Protocol (MCP) server infrastructures supporting scalable AI deployment.
- Implemented advanced observability and monitoring using LangFuse for comprehensive LLM application tracking, performance optimization, and cost management.
𝐒𝐭𝐫𝐚𝐭𝐞𝐠𝐢𝐜 𝐀𝐈 𝐈𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 & 𝐋𝐞𝐚𝐝𝐞𝐫𝐬𝐡𝐢𝐩
- Collaborated with executives and cross-functional teams to analyze complex business requirements and design tailored AI solutions.
- Maintained thought leadership in AI innovation through active participation in industry conferences, workshops, and research initiatives.
May 20212 yrs 6 mosOct 2023
Built Advance Analytics Solutions for Customer OPS
- Engineered scalable DL architectures and NLP models for a Global Pharmaceutical leader to automate Adverse Event (AE) of Drug identification in Medical Triage and Pharmacovigilance.
- Deployed Speech-to-Text (STT) and NLU frameworks to extract real-time actionable insights from patient interactions for medical safety decision support.
- Solved problems using descriptive and predictive analytics.
- Applyed robust, efficient and scalable deep learning based solutions on real-time customer data for extraction of business value through insights and recommendations.
- Replaced a lot of manual effort to bring automation wherever necessary and possible through AI.
- Automated manual medical auditing workflows through AI-driven solutions, delivering a 200% ROI while enhancing data integrity for complex regulatory compliance.- Using Deep learning based solutions to build robust models , helping to provide robust client solutions .
- Enhancing the capabilities to greater heights through working in versatile challenges in speech and text technology.
- Configured and managed multi-GPU clusters (NVIDIA A100/H100) using Distributed Data Parallel (DDP) strategies to optimize Large Language Model (LLM) training efficiency.
- Orchestrated end-to-end environment setup and resource allocation for deep learning workloads, ensuring 99.9% uptime for mission-critical Drug Safety monitoring pipelines.
- Led cross-functional teams to transition experimental AI research into "Production-First" scalable healthcare applications for real-time clinical use.
Jul 202011 mosMay 2021
Bangalore Urban, Karnataka, India
Customer Success Engineering
● Established a Quality-driven approach to delivering excellent customer satisfaction by ensuring three key principles are in place:
Understand | Empathize | Solve
● Achieved 100% customer satisfaction score in consecutive quarters.
● Able to achieve 99% SLA for resolving customer issues on connectors.
● Built data pipelines on the cloud, dbt, data flow, and model data warehousing for enterprise customers ensuring a successful transition from on-premise to cloud.
● Experience in CustomerFocused problem solving on Cloud-based ETL platform for end-to-end solutions.
● Consulting customers about the best sources to choose from based on their volume, variety and complexity of data.
● provided efficient solutions by setting up the right pipeline for the right data.
● Worked on various modern sources of data pipelines such as Redshift, Postgres, Big-query, asana, Hubspot, Facebook ads, google ads, AWS S3, cloud files etc.
● Expertise in simplifying product complexity and challenges through the abstraction of too many technical details of products and features to help customers have a smooth product experience.
Dec 20182 yrs 4 mosMar 2021
Bangalore
CellStrat is India’s leading consulting and training firm in emerging areas of Artificial Intelligence and Deep Learning. This position is an non-paid internship + academic AI research program.
At CellStrat I have learned about being a searcher of technology. I later applied my research skills for Machine Learning and Deep Learning and NLP domain.
- Expertise in end to end product building and deployment on AWS.
- Developed several NLP based solutions such as Topic Modeling, Text Summarization, Key word generation, question answering etc and deployed on production ready servers.
- Worked on research paper publication and structuring .
- Learned to develop and apply research paper based solutions on public data.
- Won several hackathons and competitions on ML and deep learning based problems.
- Experience training data scientists on applying basic to advance NLP based solutions.
- Organized and hosted many webinars on all topics around AI, DL and NLP.
Apr 20173 yrs 4 mosJul 2020
Bangalore
* Engineered ETL data pipelines with PowerCenter and IDQ transformations, processing enterprise-scale volumes to deliver 99%+ data accuracy for client analytics.
* Automated CI/CD for IDQ jobs via GitHub + Jenkins, slashing deployment time by 70% and enabling zero-downtime releases.
* Resolved complex IDQ/AWS production issues through log parsing and performance tuning, minimizing MTTR to under 2 hours.
* Built Cloud Data Quality microservices for real-time data profiling, cleansing, and observability in hybrid environments.
* Deployed ML-driven rules (bigrams, edit distance, tokenization, regex) for deduplication, null handling, and error correction on terabyte datasets.
* Engineered probabilistic fuzzy matching to correlate fields across sources, boosting entity resolution accuracy to 95%+ for master data management.
* Integrated sentiment analysis on customer logs/comms to trigger business alerts, reducing churn risks by surfacing issues proactively.
* Authored DQ training modules for Informatica University and onboarded 20+ engineers, standardizing best practices.
* Mentored teams on ETL fundamentals, accelerating ramp-up and cross-functional delivery.
Sep 20124 yrs 8 mosApr 2017
Bangalore Urban, Karnataka, India
- This role begins my journey into data, data engineering and ETL fundamentals.
- Learned, developed and gained experience in core ETL architecture, applying os ( Linux, windows) fundamentals, database concepts and file handling.
- strong experience in writing efficient and complex SQL queries for enterprise customers using Informatica Power Center.
- Developed a great understanding of end-to-end data pipelines, different source systems, and engineering best practices for data warehousing.
- Gained experience working on data warehousing, OLTP and OLAP-based systems for enterprise customers.
Education
Master of Technology - MTech, Data Analytics
Birla Institute of Technology and Science, Pilani
20162 yrs2018
Information Technology, Computer Engineering
Maulana Abul Kalam Azad University of Technology, West Bengal formerly WBUT
20084 yrs2012
Join Indrajit and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal Boss