Headline

AI Engineer II @ Flipkart | Voice AI, RAG, Agentic AI | Patent Co-Author

About

I build AI systems that make safety-critical decisions in automotive — where getting it right isn't optional. As Lead Gen AI Engineer at Bosch Global Software Technologies, I lead a cross-functional team of 8+ engineers building AI-powered products spanning Generative AI, LLM fine-tuning, functional safety (FuSA) AI, and cloud-native platforms. From HARA AI that automates Hazard Analysis per ISO 26262, to RAG platforms processing 100MB+ technical documents, to voice-enabled developer tools — I specialize in turning complex AI capabilities into production-ready systems. What I bring to the table: → Led development of HARA AI, automating safety goal identification and ASIL classification for automotive systems → Won Bosch AWS Hackathon 2025 with "Project IQ" — 90% accuracy extracting competencies from SRS docs, saving €25K+ annually → Co-authored patent on ethernet frame detunneling (No: 202541072960) → Represented Bosch at Bharat Mobility Expo 2025, showcasing Vehicle Assistant AI → Built BRICK, an advanced PathRAG document analysis platform with structural preservation → Architected serverless multi-tenant SaaS reducing onboarding time by 90% Core Skills: Python, Java, C++, Rust, TypeScript | LLM Fine-Tuning (QLoRA/LoRA/PEFT), RAG, Agentic AI, LangChain, DSPy | PyTorch, Transformers, Hugging Face | FastAPI, Django, Spring Boot, Angular, Next.js, React | AWS (Lambda/CDK), Docker, Kubernetes | gRPC, MQTT, WebSocket, CAN/DBC, VSS Let's connect! I love talking about GenAI, automotive software, and building products that matter. 📧 vepamanisaikiran@gmail.com | 🌐 www.vepamani.com

Location

Bengaluru

AI stack

claude code

codex

cursor

elevenlabs

Deepgram

Work experience

AI Engineer II at Flipkart, Jun 2026 - Present · 5 mos, Bengaluru, Karnataka, India

Lead Gen AI Engineer at Bosch Global Software Technologies, Jul 2025 - Jun 2026 · 1 yr, Bengaluru, Karnataka, India, • Leading a cross-functional team of 8+ engineers across Gen AI, automotive middleware, and cloud platform initiatives, driving technical architecture and code quality standards. • Won Bosch AWS Hackathon 2025 for "Project IQ" — an AI-powered training & knowledge management system achieving 90% accuracy in extracting competencies from SRS documents, saving €25,000+ annually. • Built HARA AI — a safety-relevant AI system automating Hazard Analysis & Risk Assessment (HARA) per ISO 26262, generating safety goals and ASIL classifications to reduce manual safety assessment effort. • Architected BRICK, an advanced document analysis platform with PathRAG and MinerU for high-precision retrieval from unstructured PDFs with layout recognition for formulas, tables, and images. • Co-authored patent on detunneling data from ethernet frames using application-specific unique IDs (Patent No: 202541072960). • Represented Bosch at Bharat Mobility Expo 2025, showcasing a Vehicle Assistant AI that generates SDV apps deployable on HMI and HPCs.

Senior Software Engineer at Bosch Global Software Technologies, Apr 2023 - Jul 2025 · 2 yrs 4 mos, Banaglore, • End-to-end architected HireStream, a recruitment OS matching associate skills against JDs using zero-shot learning, reducing manual screening by 80%. • Performed Parameter-Efficient Fine-Tuning (PEFT) on Qwen 7B+ models using QLoRA/LoRA with 4-bit quantization on limited hardware for automotive domain alignment. • Developed SDX Assistant, a voice-enabled developer tool with STT/TTS and multi-turn conversation for natural language code generation while interacting with in-vehicle hardware. • Built SDX-A2L Signal Monitor with Two-Stage AI Search (20x faster) for parsing 100MB+ A2L files and real-time ECU memory monitoring via WebSocket. • Architected a Serverless Multi-Tenant SaaS platform (XSPACE Cloud) using AWS CDK, reducing new vehicle project onboarding time by 90%. • Developed high-compliance backend microservices for Smart Connected EV platform using Java 17/Spring Boot with containerized Kubernetes deployments ensuring 99.9% uptime.

SDE II at HashedIn, Aug 2021 - Apr 2023 · 1 yr 9 mos, Bengaluru, Karnataka, India, • Built cloud-native content normalization platform for Thomson Reuters legal domain using AWS serverless architecture. • Developed annotation component in Angular that reduced document content processing time from 7 days to 4 hours. • Architected AWS serverless cloud infrastructure using Lambda, API Gateway, RDS, DynamoDB, OCR, CodePipelines, and CloudFormation. • Collaborated with US legal domain stakeholders for requirement gathering and defining technical user stories.

Associate Software Engineer at ZapCom Group Inc, Feb 2020 - Jul 2021 · 1 yr 6 mos, Bangaon, West Bengal, India, • Built REST APIs using Python/Django DRF and migrated database from Kinvey to Django models. • Implemented FCM push notifications and Celery-based async task processing in Python. • Created and maintained AWS instances for daily review extractions with NLP scoring using spaCy. • Built Content Management service interface handling 1000+ client devices in WSO2. • Redesigned 70% of UI controlling user flow architecture, improving user experience. • Wrote automation test scripts in Python using Selenium, reducing manual testing effort by 35%.

Intern at ZapCom Group Inc, Jan 2019 - Jan 2020 · 1 yr 1 mo, Bangalore, • Redesigned Zapcom portal with responsive landing pages in HTML/CSS, improving navigation and cross-browser compatibility. • Built scene classification model using Places365 dataset for image recognition. • Developed web scraping pipeline in Java (Selenium) and Python (Scrapy) to extract 1000+ reviews from 540+ client hotels. • Integrated Facebook Developer API to automate review extraction for 100+ client hotels.

Education

Bachelor of Technology, Computer Science, JNTU Anantapur