AI ENGINEER — SURAT, IN

Intelligent systems, built in production

why i chose ai engineering

because the future thinks

01 / Profile

Engineer-

01Frame
The problem

Scope, data, and a success metric — defined before any code.

02Wire
The system

Agents, retrieval, and tools — wired into one pipeline.

03Ship
The release

Evals, logs, production — measured, not demoed.

Frame
Wire
Ship
Run it again

AI Engineer specializing in generative AI, LLM agents, and RAG systems.

I build and ship production AI workflows — fine-tuning language models, wiring agent systems with tool use, and deploying retrieval-backed pipelines with PyTorch, LangChain, and FastAPI.

My work runs the full loop: framing the problem, integrating the APIs, evaluating the outputs, and getting it into production — from multi-step agent automation to reproducible image-generation workflows.

LocationSurat, Gujarat, IN — 21.17°N 72.83°E
EducationB.Tech CSE (AI), Parul University — 2023–27
02 / Experience

Thinking-

AI EngineerSeepossible

production ai systems · llm apis · agent workflows

  • Building and deploying production AI systems on LLM APIs — agent workflows, tool use, and retrieval-backed pipelines.
  • Working across the full loop from problem framing to API integration, evaluation, and deployment.
Nevil Choksi — Surat, IndiaAug 2026 — Present
By the numbers

Measured-

Every system below shipped with a number attached. Not demos — production workflows with before/after deltas, measured across real team usage.

Four deployments, four deltas
03 / Projects

Built, measured,
shipped.

Four production systems — every one shipped with a number attached. Not demos: indexed codebases, agent memory, and diffusion pipelines running for real teams.

Codebase Index
0files indexed
84k symbols3 MCP tools<90s full index
Codebase Intelligence · MCP

Linear-Canal

A context intelligence layer for AI coding agents — full structural awareness before a single file is touched.

linear-canal.com
Memory Tiers
Hot storeRedis
SemanticQdrant
PersistentPostgres
<200msretrieval across all tiers
Multi-Agent Memory · Go

Context Forge

Multi-agent context management with layered memory — hot store, semantic index, durable persistence.

Training Run
E1
E2
E3
E4
E5
bestE6
−40%training time · mixed precision
Diffusion · Training Pipeline

LoRA Training Pipeline

End-to-end domain-specific image generation, deployed as a concurrent inference service.

Belief ledger
B-114user prefers dark ui
0.97
B-098deploys via wrangler
0.91
B-087pnpm over npm
0.88
recall"which package manager?"→ B-087
3beliefs verified · chain intact, tied to source
Memory infrastructure · alpha

Helix — Memory System

Turns agent sessions into durable beliefs tied to their source, time, and history — memory you can audit.

04 / Skills

Tools-

01

Machine Learning

CORE
Deep LearningComputer VisionNLPRAG SystemsModel Fine-tuning
02

Frameworks & Tools

DAILY
PyTorchTensorFlowLangChainHugging FaceStable DiffusionComfyUI
03

Development

BUILD
PythonJavaScriptReactFastAPISQLDockerGitPostgreSQLVector DBs
04

Automation

OPS
n8nWeb ScrapingAPI Integration
05 / Education

Study-

B.Tech — Computer Science Engineering

AI SPECIALIZATION
Parul University · Vadodara, Gujarat
2023 – 2027 · Expected 2027
20232027
YR 4/4

final year — in progress

Coursework weighted toward applied machine learning — from systems fundamentals to deep learning and computer vision.

Machine LearningDeep LearningComputer VisionData StructuresOperating SystemsDBMS · Cloud

Certifications

Oracle Cloud Infrastructure 2025 Gen AI ProfessionalOCI
Deloitte Australia Data Analytics SimulationFORAGE
IBM AI SkillBuilderIBM
06 / Contact

Let's build
something intelligent.