Open to full-time AI Engineer roles · Hyderabad

G Sudarshan
Sastry

Full-stack AI engineer building LLM, RAG and voice-agent systems that run in production.

I take AI products from requirement to deployment: FastAPI and Next.js services, vector retrieval, real-time voice, and the multi-cloud infrastructure underneath. First author at IEEE IGARSS 2025.

Portrait of G Sudarshan Sastry
CurrentlyAI Engineer (part-time), Anju Siima Technologies
Shipped
14+
LLM / RAG systems for 9 client accounts
Research
IEEE
First-author paper, IGARSS 2025
Worked with
  • Anju Siima Technologies
  • Freedom With AI
  • Deep Algorithms
  • IIT Mandi
  • AKOSCOWL
Experience

Where I've worked

From firmware and robotics labs to running client AI systems in production.

Sep 2026 – Present
Remote · Part-time
Current
AS

AI Engineer

Anju Siima Technologies Pvt. Ltd · Product Development
  • Building a multi-tenant AI voice-calling platform: bots with versioned prompts, phone and browser test calls, a call queue with retries and circuit breakers, campaigns, recordings, transcripts and LLM post-call analysis (FastAPI, MongoDB, LiveKit, Docker).
  • Made every speech and model vendor a plugin (Sarvam, Deepgram, ElevenLabs, Murf, Fish Audio, Gemini, OpenAI) and benchmarked 12 TTS models, so each bot can switch providers on latency, cost and language quality.
  • Shipped multilingual calling agents (English, Hindi, Telugu) with tool calls for demo booking on Google Calendar, scheduled callbacks and hand-off to a human.
Nov 2025 – Jul 2026
Remote
FW

Associate AI Developer

Joined as AI Intern (Nov 2025), converted to Associate AI Developer (Jan 2026), full-time from Apr 2026.
  • Delivered 14+ LLM/RAG systems across 9 client accounts with provider-agnostic LLM routing (retry and fallback), vector retrieval, FastAPI and Next.js services and n8n event orchestration, owning each from requirement to deployment.
  • Productised a webinar and call attribution platform (Next.js, PostgreSQL/JSONB) that reconciles 4 mismatched export sources into one schema across 7 migrations; one codebase redeployed to 5 tenants.
  • Recovered ~37,000 validated CRM contacts and 61 automations from a broken migration by mapping an undocumented internal API.
  • Shipped an AI tool suite on Cloud Run (124 commits): prompt enhancer, AI tool finder, job-prep assistant, a Chrome extension and a Claude-based assignment auto-grader.
  • Operated 16+ Dockerised services on a six-VM AWS/GCP/OCI fleet (43 Dockerfile/Compose artifacts) with SSH runbooks; distilled a 49,737-message mail corpus into a ~5,000-pair support knowledge base; ran knowledge transfer for ~12 colleagues.
Mar 2025 – Sep 2025
Hyderabad
DA

Software Engineer

Intern from Mar 2025, Software Engineer from May 2025.
  • Built a DDoS detection path with packet-to-flow aggregation computing per-source packet rate, source-IP entropy and protocol-mix features, evaluated against tunable thresholds for volumetric and protocol-layer attacks.
  • Implemented TLS fingerprinting from ClientHello metadata (cipher suites, extensions, version) so detections key on client-stack identity rather than source IP; owned feature specs and acceptance tests with QA.
Dec 2024 – Mar 2025
Mandi
IIT

Research Intern

CAIR Lab, IIT Mandi
  • Wrote STM32 BLDC motor firmware in C with timer-driven PWM commutation and closed-loop speed control, from bare-metal bring-up to bench validation.
  • Fused multi-sensor gas readings on Arduino using per-sensor calibration curves, drift compensation and noise filtering; wrote the firmware docs and test plans the lab adopted.
Feb 2024 – Sep 2024
Hyderabad
AK

AI-ML Intern

AKOSCOWL DOT INVENTIONS Pvt. Ltd.
  • Designed drone power and control electronics in KiCad and built a ROS/Gazebo model (URDF airframe, sensor plugins, flight-controller-in-the-loop).
  • Developed obstacle-avoidance planners emitting MAVLink waypoints through QGroundControl mission flows, validated in simulation before flight testing.
Earlier internships (7) +
  • Machine Learning Intern — Bharat InternDec 2023 – Jan 2024House-price prediction (XGBoost), TF-IDF movie recommender, Iris classification.
  • Python Development Intern — CodeClauseJul – Aug 2023Voice assistant, plagiarism checker, URL shortener, music player. Letter of Recommendation.
  • AI with Python Intern — CoincentMay – Jun 2023Landmark-detection project after a two-month AI with Python program.
  • Python Development Virtual Intern — Coders Cave1 monthMusic player, calculator and two other GUI apps.
  • Data Engineering Virtual InternshipAICTE-EduSkills × AWS Academy · Sep – Nov 2023
  • Intelligent Automation Virtual InternshipAICTE-EduSkills × SS&C Blue Prism · May – Jul 2023
  • AI-ML Virtual InternshipAICTE-EduSkills × AWS Academy · Jul – Sep 2022
Projects

Things I've built

LLM tooling, automation and local-first AI, built end to end.

Research

Published work

IEEE IGARSS 2025 · First author

Obstacle Avoidance in Drones Using Deep Reinforcement Learning

DQL and DDPG agents trained in Unity3D compare 2D LiDAR against depth estimation for fixed-wing UAV obstacle avoidance. LiDAR won for real-time control, and the paper proposes hybrid sensing for production.

G Sudarshan Sastry, Ravi Boda, Subhranginee Das, K. Arun Kumar, Mousmi Ajay Chaurasia
DQL · DDPGDeep RL agents compared
LiDARwon for real-time control
IGARSS 2025IEEE Geoscience and Remote Sensing Symposium
Toolkit

What I work with

AI / LLM

RAGFAISSQdrantpgvectorLangChainLLM routingLLM-as-judge evalsGeminiOpenAIClaudeOllamaPyTorchTensorFlowscikit-learn

Backend & web

FastAPIFlaskDjangoREST APIsNext.jsReactNode.jsPrismaWebSocketsWebRTCPipecat

Infra & data

DockerLinuxAWSGCPOCICloud RunVercelPostgreSQLSupabaseRedisMongoDBSQLiteGit

Languages

PythonJavaScriptTypeScriptCJavaRustSQL

Automation

n8nPabblyBlue Prism RPAWebhooksCron

Robotics & embedded

ROSGazeboUnity3DKiCadSTM32MATLAB
Credentials

Certifications & education

Every certificate I hold. Where a public verification page exists, it's linked.

Courses & training

MongoDB University · 2023
Python Developer PathDocument ModelGetting Started with AtlasConnecting to MongoDBConnecting in PythonCRUD: Insert & FindCRUD: Replace & DeleteCRUD: Modifying ResultsCRUD in PythonAggregationAggregation in PythonIndexesData ModelingTransactionsAtlas Search
SS&C Blue Prism · 2023
Foundation TrainingAdvanced Data ItemsAdvanced Attribute MatchingProcess TemplatesCredentials & Credentials ManagerAssociate Developer Next StepsCertification Readiness Webinar
Programs & training
Amazon ML Summer School 2023AI with Python — CoincentTensorFlow for Beginners — SimplilearnAndroid Fundamentals Master Class — Perfect Plan B
Internship certificates
AICTE-EduSkills × AWS: AI-MLAICTE-EduSkills × AWS: Data EngineeringAICTE-EduSkills × Blue Prism: Intelligent AutomationCodeClauseBharat InternCoincentCoders Cave

Recognition

First-author IEEE paperIGARSS 2025
1st place, JHUB-JNTUH Regional HackathonJune 2022
Official Coding MentorDept. of ECE, KL University
Four letters of recommendationIncluding the ECE Head of Department

Education

B.Tech, Electronics & Communication Engineering

KL University, Hyderabad · Specialization in Data Science & Big Data Analytics · Aug 2021 – Aug 2025

8.59CGPA / 10
Contact

Looking for an AI engineer?
Let's talk.

Open to full-time roles in Hyderabad or remote.