Aprameya C V

AI engineer building agentic and generative AI systems, from model to production.

I work across applied AI end to end: agent design, retrieval, fine-tuning and model serving, and the cloud infrastructure that runs them. I like problems with a measurable answer, so every project here comes with its numbers and its limits.

Role
AI engineer
Focus
Agents, retrieval, inference
Open to
Applied AI, forward-deployed
Portrait of Aprameya C V
Scroll

About

Who I am

An engineer who likes questions with a measurable answer.

I am an AI engineer based in Bengaluru. I started as an engineering intern in January 2025, moved into a full-time role that summer, and now lead technical scoping for customer AI engagements across EMEA.

Most of my day is the gap between a prototype and something a customer can rely on: scoping the problem in the first conversation, building the agents, retrieval and serving layers, and deploying them on OCI. I have built systems for government and tax-authority clients, including multilingual assistants and image-matching APIs that run on a CPU-only machine.

Outside client work I run small, controlled studies on the same questions: how many passages an agent should retrieve, which serving flag actually helps, when a cheap model should hand off to a strong one. Each is published here with its numbers and its limits.

I studied electronics at JSS Science and Technology University, where I published papers on seismic signal analysis, IoT intrusion detection and satellite link design, and hold a design patent. That is where the habit comes from: state the question, fix the variables, report the interval.

01 / Skills

What I work with

The core of what I build with, grouped by the layer each sits in.

Generative AI and AI Engineering

  • RAG and hybrid search
  • LLM evaluation
  • Fine-tuning (LoRA)
  • Multimodal and voice AI
  • PyTorch
  • Hugging Face

Shown in HBC-RAG · Pramana · CG-BI

Agentic AI

  • LangGraph
  • LangChain
  • OpenAI SDK
  • MCP
  • ADK

Shown in HBC-RAG · Experience

AI Infrastructure

  • vLLM
  • CUDA and NVIDIA GPUs
  • Speculative decoding
  • Model serving

Shown in SpecLane

Cloud and DevOps

  • Oracle Cloud (OCI)
  • AWS
  • Google Cloud
  • Kubernetes
  • Docker
  • Terraform

Shown in Experience · Demand forecasting

Programming

  • Python
  • TypeScript
  • SQL
  • FastAPI

Shown in HBC-RAG · Demand forecasting

02 / Experience

Oracle

Three roles in under two years, from intern to technical scoping lead.

0+ hHours savedInternal AI tools, estimated
0–60+ hSaved per engagementDeployable solution packs for RAG, agents, voice, MCP and guardrails
0M+Tickets searchableServiceNow suite, answers in under 7 seconds
0K+Images matchedTrademark platform, thousands of checks a week
Jun 2026 to present

Oracle / Cloud Engineering, AI (L2)

  • Technical scoping lead and point of contact on EMEA engagements across GPU serving, guardrails, voice AI, MCP and agentic automation.
  • Built a three-agent ServiceNow suite for a public-sector client: natural-language search over 1M+ tickets in under 7 seconds, hybrid search over 500K+ multilingual articles, and ticket creation in under a minute.
  • Built a multilingual legal and tax RAG assistant for a national tax authority, so citizens can ask questions in their own language.
  • Delivered OCI infrastructure end to end: landing zones, networking, storage, IAM, VM migration and disaster recovery, automated with Terraform.
  • Built deployable solution packs for RAG, agents, voice, MCP, guardrails and agent memory, and drove Codex adoption for 80+ people who built 40 to 50 reusable assets.
Jul 2025 to May 2026

Oracle / Cloud Solution Engineer (L1)

  • Built a trademark platform for a government ministry, exposed as two APIs. Similarity search embeds 500K+ trademark images with OpenCLIP into Oracle Database vector search and blends lexical and semantic matching, bringing latency from about 300 seconds to under 10 on a CPU-only machine.
  • Built the generation side of the platform: Flux for text-to-image and a custom ComfyUI workflow for editing on a 24 GB NVIDIA GPU, fitted with attention slicing and offloading, with Arabic support and a content moderation filter. Showcased live at GITEX.
  • Primary developer of internal AI tools, saving an estimated 1,100+ hours: an agentic security-audit pipeline used on 100+ projects, a document generator, a RAG knowledge workspace over 1,000+ documents for 100+ users with single sign-on, and an Open Landing Zone configuration generator that cut setup from 3 to 5 days to hours.
  • Delivered OCI infrastructure on customer engagements, including landing zones, IAM, networking, load balancing and 40+ VMs, and migrated a 24 TB workload with RackWare, the largest the team had handled.
Jan 2025 to Jul 2025

Oracle / Project Intern

  • Built the Solution Document Generator on LangGraph and OKE, contributing to 50+ documents and 150+ person-hours saved.
  • Built a security review agent used to review 50 to 60 production go-live projects, saving 100+ person-hours.

Project results

0xLLM serving throughput from speculative decodingSpecLane
0%right clause found, up from 47%, in half the timeHBC-RAG
0%critical defects caught at 47% of the strong model's costCG-BI
0 ptsaverage gap between stated confidence and accuracyPramana

06 / Contact

Building something that needs to work?

I am open to applied AI and forward-deployed engineering roles. Email is the fastest way to reach me.

Email copied