Data Engineering
Large-scale ETL and streaming pipelines on Spark, Kafka, and the cloud, plus Teradata & Snowflake warehousing with query tuning and orchestration at production scale.
Data, AI & Software Engineer · Portfolio 2026
AkshayaaPrabakar
Data engineer building large-scale pipelines, warehousing, and cloud infrastructure, and a software & AI engineer bringing LLMs, RAG, and agentic systems to production on top of them. M.S. Computer Science, NYU 2026.
About
Data systems first — intelligence on top.
I build the data foundations that production systems run on, large-scale ETL and streaming pipelines, warehousing, and cloud infrastructure, and layer applied AI on top, from LLMs and RAG to multi-agent workflows. I work across the full lifecycle: data preparation and retrieval, model integration, and rigorous evaluation, turning ideas into reliable, real-world software.
3+Years building data & AI systems
~300Grad students mentored as Head TA
~40KDocuments processed for LLM training
1Startup co-founded (NYU Leslie Entrepreneurial Institute)
Applied GenAI & LLMs
RAG over vector databases with automated retrieval evaluation, LLM fine-tuning and Domain-Adaptive Pre-Training, prompt engineering, and multi-modal generation.
Agentic Systems
Multi-agent orchestration with LangChain, LangGraph, and LangSmith, with tool-using agents (Claude, Hugging Face, Gemini) for multi-step automation.
Trustworthy ML & Governance
Automated validation gates, data quality frameworks, PII protection, and audit trails, aligned to NIST AI risk frameworks and ISO 42001.
Experience
Three years across data, AI & engineering.
- Jan 2026 – Present
Software Engineer (Intern to Full Time)
CEARTscore · New York, NY
- Built a RAG system over vector databases with embedding workflows, and multi-agent LLM workflows that automated multi-step document analysis and content generation, cutting manual analysis effort significantly.
- Designed and automated a multi-source data pipeline on Azure processing ~1,000 documents, with PII reduction, validation, secure storage, and preprocessing for model training.
- Prepared and processed a corpus of ~40K documents for domain-specific LLM fine-tuning and Domain-Adaptive Pre-Training.
- Built an automated evaluation harness that runs on each pipeline execution, scoring RAG retrieval against known-good answers and surfacing what passed, what failed, and the retrieved output beside the expected one.
- Jun 2025 – Aug 2025
Software Engineering Intern
Block Convey · New York, NY
- Built an end-to-end synthetic data generator with multi-stage PII masking; a core contribution to the company’s patent portfolio.
- Re-engineered a monolithic pipeline into a parallel, agent-based system on GCP, roughly doubling processing throughput (~50% faster) via distributed workflow handling, validation, and monitoring.
- Developed a NIST-aligned data validation and quality framework that reduced manual review and reporting time by ~80% through automated pre-validation.
- May 2025 – May 2026
Graduate Teaching Assistant, Big Data (CS-GY-6513)
New York University · New York, NY
- Head TA for the course; mentored nearly 300 graduate students in Hadoop, Spark, Dask, Kafka, and NoSQL, held TA hours and clarified queries, evaluated assignments with targeted feedback, and guided 200+ project teams from proposal to final demonstration.
- Oct 2022 – Jul 2024
Data Engineering Associate
Accenture Solutions Pvt. Ltd. · Chennai, India
- Optimized data transformation processes using Ab Initio and Informatica PowerCenter, improving data accuracy and ETL performance by ~40% through advanced Teradata query tuning, and streamlined dozens of Control-M jobs to enhance ETL workflows and reduce processing errors, ensuring reliable validation, migration, and testing.
- Designed and refined stored procedures, views, and tables in Teradata and Snowflake, improving query performance to support data-driven decision-making.
Capabilities
A full-stack data & AI toolkit.
From large-scale data pipelines and cloud infrastructure through to the generative AI and agents that run on top of them.
Data Engineering
- SQL / PL-SQL
- Apache Spark
- PySpark
- Hadoop
- Kafka
- Dask
- Ab Initio
- Informatica PowerCenter
- Control-M
- Teradata
- Snowflake
Cloud & DevOps
- AWS (S3, EC2, Lambda)
- Azure
- GCP
- Docker
- Kubernetes
- Jenkins
- CI/CD
- Git / GitHub
AI & GenAI
- Large Language Models
- RAG
- LLM Fine-Tuning
- Domain-Adaptive Pre-Training
- Prompt Engineering
- Evaluation Harnesses
- Multi-Modal Generation
- Embeddings
- Vector Databases
Agentic Systems
- LangChain
- LangGraph
- LangSmith
- Multi-Agent Orchestration
- Tool-Using Agents
- Claude
- Hugging Face
- Gemini
Machine Learning
- TensorFlow
- PyTorch
- Model Evaluation
- Synthetic Data Generation
Languages & Governance
- Python
- SQL
- Bash / Unix
- NIST AI Risk
- ISO 42001
- PII Protection
- SC-900
Selected Work
Things I’ve built.
Research-to-Podcast Pipeline
An agentic pipeline that runs deep web research through Tavily with Perplexity fallback, generates a narrated analysis script via the Claude API, and auto-triggers ElevenLabs TTS to produce finished podcasts delivered to client teams. End to end with no human in the loop. The hard part was defining what “good” means for output nobody can grade.
Claude APITavilyElevenLabsMulti-AgentProduction
AI Context Engine
An MCP-based system that routes AI-proposed code changes through simulated domain-expert personas, tiering scrutiny by risk and catching security, compliance, and team-standard violations before a PR is ever opened. Validated through 70+ B2B customer discovery interviews.
MCPMulti-AgentLangGraphRisk TieringGovernance
Eclipse
A policy-controlled treasury for autonomous AI agents, with deterministic spending rules and on-chain settlement backed by verifiable audit trails.
Autonomous AgentsOn-Chain SettlementAudit Trails
Enterprise ETL & Warehouse Optimization
Large-scale ETL pipelines feeding Teradata and Snowflake warehouses, built with Ab Initio, Informatica PowerCenter, and Control-M orchestration, with query tuning and stored-procedure design to keep production data processing fast and reliable.
TeradataSnowflakeAb InitioInformaticaControl-M
Aug 2024 – May 2026
M.S. in Computer Science
New York University · United States
GPA 3.5 / 4Aug 2018 – May 2022
B.E. in Computer Science & Engineering
Anna University · Chennai, India
8.2 / 10Certification
SC-900, Security, Compliance & Identity
Let’s build
Have an idea worth engineering?
I’m open to roles and collaborations in data engineering — as well as AI and software engineering. The fastest way to reach me is email.
ap9013@nyu.edu