Dileep Kumar Varada
Open to work1 years experienceAI Engineer
AI Engineer with R&D experience in autonomous driving, computer vision, and multimodal sensor fusion.
Standing Rank
Rank Not Available
Developer Badges
No badges earned yet
Skills & Technologies
33 skillsWork Experience
Research & Development Intern – Autonomous Driving & AI Systems
Bosch Global Software Technologies (BGSW)
- Built AutonomaVLA, a tri-modal (camera + LiDAR + radar) BEV fusion framework with a quantized Qwen2-0.5B backbone (QLoRA), achieving Driving Score 63.7 / Success Rate 63.6% on Bench2Drive (55 closed-loop routes)
- Designed a BEV fusion pipeline with SE-Attention and a Qwen2-0.5B backbone fine-tuned via 4-bit NF4 QLoRA, optimized for deployment on a single 6GB GPU.
- Designed a Cartesian GRU trajectory decoder that resolved lateral collapse present in the earlier polar-coordinate decoder, producing multi-hypothesis (K=3) ego-trajectory predictions
Projects
AutonomaVLA – Tri-Modal VLA Driving Framework
- Tri-modal camera+LiDAR+radar BEV fusion with SE-Attention and radar Doppler features, fine-tuned via 4-bit NF4 QLoRA on a Qwen2-0.5B backbone for single-GPU (6GB) deployment
- Achieved DS 63.7 / SR 63.6% on Bench2Drive (55 routes); fixed lateral-collapse failure mode via Cartesian (vs. polar) GRU decoder design.
AI-First CRM HCP Module – Full Stack Agentic Application
- Built a LangGraph ReAct agent with 5 custom tools for automated CRM workflows, migrated from a decommissioned Groq model to llama-3.1-8b-instant with zero downtime
- Designed a LangGraph-based ReAct agent powered by Groq LLM inference with five custom tools for automated CRM data retrieval and workflow actions, with end-to-end state management and REST API integration.
Secure Serverless Data Pipeline Orchestration
- Architected a fault-tolerant, event-driven data pipeline using AWS Lambda, Step Functions, and Amazon S3 for scalable serverless processing.
- Optimized performance through cold-start tuning and parallel Lambda fan-out across S3 partitions, achieving 40% lower processing latency compared to a monolithic baseline.
Leaderboard Standings
Leaderboard Position Pending
Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.
Assessment Highlights
Assessments Not Completed
Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.
AI Collaboration Score
AI Collaboration Score Pending
Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.
Role Compatibility Profile
Role Compatibility Analysis Pending
Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.
Achievements
Single-Frame Ego-Trajectory Prediction with Structured Motion Priors
Single-Frame Ego-Trajectory Prediction with Structured Motion Priors, IC2E3 2026, Accepted
A Tri-Modal Vision-Language-Action Framework with Cartesian Trajectory Decoding
A Tri-Modal Vision-Language-Action Framework with Cartesian Trajectory Decoding, IC3AI 2026, Accepted
About Details
Professional Bio
AI Engineer with hands-on R&D experience in autonomous driving, computer vision, and multimodal sensor fusion. Skilled in deep learning, NLP, agentic AI systems, and full-stack AI application development.
M.Tech in Computer Engineering in Computer Engineering
National Institute of Technology Kurukshetra, Haryana (2024 - 2026)
B.Tech in Computer Science & Engineering in Computer Science & Engineering
Sri Krishnadevaraya University, Anantapur (2019 - 2023)