Swapnil Shindemeshram

CEO

Bengaluru, Karnataka, India7 yrs 6 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Led architecture for high-scale systems at CleverTap.
Reduced integration time from 1 month to 1 week.
Developed a patented real-time evaluation system.

Stackforce AI infers this person is a SaaS architect specializing in scalable distributed systems and GenAI solutions.

Contact

Skills

Core Skills

Distributed SystemsDatabase ManagementGenaiBackend DevelopmentFull-stack DevelopmentApi DevelopmentMachine LearningNlp

Other Skills

JavaPythonScyllaDBPostgreSQLElasticsearchSystems DesignTechnical StrategypgvectorMongoDBRedisAlgorithmsHigh-Level DesignLow-Level DesignCode ReviewDesign Patterns

About

I design and build systems that operate at the intersection of scale, reliability, and product impact. At CleverTap (~400M MAU), I've led the architecture of systems processing 1B+ messages/month, built a GenAI content assistant (RAG pipeline with vector search), and re-architected the recommendation engine to handle 20M-row catalogs at 10K req/s. I hold a patent for automated real-time evaluation of unstructured content (No. 405196). What I care about: Making the right database choice, not the default one — I've evaluated and deployed ScyllaDB, pgvector, Elasticsearch, MongoDB, and Redis for distinct workload profiles across the platform. Designing for the failure mode, not the happy path — exactly-once delivery at p95 5ms, multi-channel fallback within SLA windows, and priority-ordered transactional messaging. Shipping with leverage — the best systems I've built reduced org-wide integration time from 1 month to 1 week, and unlocked $500K+ ARR in their first year. Previously: Built production NER/NLP systems and full-stack platforms at YouBotics. Stack: Java | Python | Rust | Kafka | ScyllaDB | PostgreSQL | Elasticsearch | Redis | AWS | GCP Open to Staff/Principal Engineer roles at companies solving meaningful technical challenges at scale.

Experience

7 yrs 6 mos

Total Experience

3 yrs 9 mos

Average Tenure

4 yrs 9 mos

Current Experience

Clevertap

4 roles

Senior Manager

Promoted

Jul 2025 – Present · 10 mos

Leading architecture and technical strategy for plaform-scale systems at CleverTap (~400M MAU).
Drivin technical design across recommendations, messaging nfrastructure, and GenAI initiatives. Authored techncal design docs for 3 major platform initiatives anddrove database selection through hands-on POCs — mathing workload characteristics to storage engines acrss the platform.
Key systems owned: GenAI content asistant (RAG), recommendation engine (20M rows, 10K eq/s), cross-device Inbox (~400M msgs/month, peak 1B, Reminders platform ($500K+ ARR), and transactionalmessaging with multi-channel fallback.
Mentoring enineers on system design and architectural decision-mking. Reduced org-wide dependency integration time fom 1 month to 1 week through a shared framework adoped across all teams.

JavaPythonScyllaDBPostgreSQLElasticsearchGenAI+4

Team Lead backend Engineer

Promoted

Jan 2023 – Present · 3 yrs 4 mos

Designed the architecture and built the backend for a GenAI content assistant (RAG): Chunked campaign data, generated embeddings via OpenAI, and indexed in Elasticsearch for vector similarity retrieval. LLM generates channel-aware copy (Push, WhatsApp, Email, Web Push) served via API.
Re-architected the recommendation engine to 2x catalog scale and 10K req/s: Scaled from 10M rows / 20 columns to 20M rows / 200 columns. Built serving layer supporting 4 recommendation strategies via pgvector. Led technical design across a cross-functional team of 8 engineers.
Designed the Reminders platform, driving $500K+ ARR in year one: Built time/behavior-triggered scheduling for automated campaigns. P95 of 5ms for online execution with exactly-once delivery.
Architected cross-device Inbox at ~400M msgs/month (peak 1B): Chose ScyllaDB for high-throughput, low-latency message storage. Drove execution across 2 teams to ship in ~2 months.
Designed transactional messaging with multi-channel fallback: Priority-ordered delivery with automatic failover within SLA window. P95 under 5 seconds end-to-end.
Improved reporting scalability 10x (100 to 2K concurrent campaigns): Redesigned query execution with workload queuing and new fan-out model. Received Functional Award (Q3 2022).
Designed and built WhatsApp BSP integration pipeline enabling CleverTap to become a certified WhatsApp Business Solution Provider.
Built graph-based dependency deletion framework for recursive entity resolution, adopted across 3+ teams.

JavaElasticsearchMongoDBRedisAlgorithmsHigh-Level Design+3

Senior Software Engineer

Promoted

Jan 2022 – Dec 2022 · 11 mos

Enhanced CleverTap's reporting code and recommendation system, achieving significant speed and efficiency improvements.
Took ownership of features like Recommendations Personalisation, Catalog Personalisation, Campaign Reports, Bulletins, and Reminders.
Improved report delivery speed by 177% and reduced database queries by 80%.

JavaCode ReviewSystems DesignDesign PatternsBackend DevelopmentDatabase Management

Software Engineer

Jun 2021 – Dec 2021 · 6 mos

Revamped WhatsApp as a channel flow, laying the groundwork for CleverTap's WhatsApp business solution provider.
Quickly resolved incorrect segmentation issue, enhancing user experience and efficiency

Youbotics

2 roles

Full Stack Engineer

Feb 2020 – May 2021 · 1 yr 3 mos

Full-stack platform engineering — designed and built REST APIs (Django, FastAPI), React/TypeScript frontend, and CI/CD pipelines (Bitbucket + GCP), reducing deployment cycles by 60%+.
Architected end-to-end application infrastructure on Google Cloud Platform — Kubernetes Engine, Cloud SQL, IAM, Cloud Storage, and Compute Engine. Owned the full lifecycle from design through production deployment.

DjangoFastAPIReactTypeScriptCI/CDFull-Stack Development+1

Machine Learning Engineer

Aug 2018 – Feb 2020 · 1 yr 6 mos

Built production NER system for automated resume parsing — LSTM-based model with end-to-end ML pipeline (data labeling → training → FastAPI serving). Deployed as the core screening service for the hiring platform.
Created machine learning model to evaluate technical responses based on model answers using NLP algorithms (transformer-based). Built the evaluation pipeline from data collection through model serving.

NLPMachine LearningFastAPI