AI / LLM ENGINEER · USA

I build
intelligencethat holds
up inthe real world.

From retrieval systems to production ML infrastructure, I turn emerging AI capability into dependable products at scale.

Explore projects ↘View résumé →Download PDF ↓Download Word ↓
YOUR FRIENDLY GUIDE

Good to see you here.

Tap my avatar for a shortcut.
RETRIEVAL SYSTEMS✦MODEL OPTIMIZATION✦AI PLATFORMS✦RETRIEVAL SYSTEMS✦MODEL OPTIMIZATION✦AI PLATFORMS✦
3+Years engineering
production systems
35%Lower inference
latency
RAGSemantic retrieval
at catalog scale
18%Increase in
user engagement
How to read the career highlights

Figures are reported in the résumé PDF. The 35% latency reduction and 18% user engagement increase lack published baselines, workloads and evaluation windows. They are reported outcomes, not reproducible benchmarks. Experience figures below have the same source; request, uptime and resolution definitions are not published. Company audience size is not evidence of this system's measured reach.

01SELECTED SYSTEMS

Proof over
promises.

Representative production work. Client-sensitive implementation details are intentionally abstracted.

01
Recommendation intelligence

Semantic discovery at streaming scale

A RAG-based content discovery architecture pairing vector retrieval with tuned generation for more relevant, explainable recommendations.

Explore this system : Semantic discovery at streaming scale
DECISION NOTESelected RAG over end-to-end fine-tuning to balance retrieval quality, operating cost, and the pace of catalog change.
+30%discovery accuracy
Metric context & source

The PDF reports an improvement in content discovery accuracy. It does not define the relevance metric or say whether this is relative change or percentage points.

Baseline, sample size, evaluation window and experiment design are not published. The 200M+ audience figure is résumé-reported scale context; measured reach attributable to this system is not established here.

Source: résumé PDF

CONCEPTUAL FLOW

  1. Content metadata
  2. Vector retrieval
  3. Grounded generation
  4. Recommendations
RAGPineconeFAISSFastAPIKubernetes
02
NLP · financial systems

Real-time fraud intelligence

A production fraud detection system combining BERT language signals with XGBoost decisioning for high-volume banking workflows.

Explore this system : Real-time fraud intelligence
DECISION NOTETurned unstructured transaction context into actionable risk signals while preserving a path for review and iteration.
$2M+annual savings
Metric context & source

Annual savings is a business outcome reported in the PDF alongside a 40% reduction in fraudulent transactions; it is not a model accuracy score.

The loss baseline, annualization method, review costs and attribution method are not published. This is a reported system outcome, not a claim of independently verified personal savings.

Source: résumé PDF

CONCEPTUAL FLOW

  1. Transaction context
  2. BERT signals
  3. XGBoost scoring
  4. Risk review
BERTXGBoostPythonAWSMLOps
03
Applied language AI

Conversational support engine

A transformer-powered support experience built to resolve common requests quickly and hand off complex conversations cleanly.

Explore this system : Conversational support engine
DECISION NOTEDesigned around resolution—not novelty—with intent routing, reliable fallbacks, and measurable customer outcomes.
50K+daily interactions
Metric context & source

Interaction volume describes workload, not unique users or resolved conversations. The PDF separately reports an 85% resolution rate.

The counting unit, observation window, resolution definition and baseline are not published. Volume and resolution should not be combined into an inferred count of successful cases.

Source: résumé PDF

CONCEPTUAL FLOW

  1. Customer request
  2. Intent routing
  3. Response or fallback
  4. Support handoff
RasaTransformersNLPREST APIsMonitoring

INTERACTIVE EXPERIMENT / 01

Follow the evidence.

Load a sample incident, inspect its logs, and run a small diagnostic experiment.

Synthetic samples · Rules run in your browser · No AI model or network request. Findings are hypotheses for human review.

You can edit the synthetic logs. Avoid pasting personal information or production secrets. Maximum 12,000 characters.

What happened, and what supports it?

Run the analysis to connect a possible failure mode to exact log lines and a next investigation step.

How this experiment works

The analyzer checks for database pool exhaustion, connection timeouts and upstream rate limits. It quotes matching lines and offers fixed investigation guidance. Editing or changing samples clears the previous result. It does not execute log content, infer causality, or change any system.

Sairam Bodapothula
BASED IN THE U.S. · WORKING GLOBALLY
02ABOUT

I care about the space between a promising model and a product people can trust.

I’m an AI/LLM engineer with 3+ years building production NLP and machine learning systems. My work spans model selection, retrieval design, inference optimization, and production operations.

I’m most useful where the problem is still a little messy: when accuracy, latency, cost, and user experience all need a seat at the same table.

Download résumé ↘
03CAPABILITIES

Built across the
whole system.

01

Intelligence

  • PyTorch
  • TensorFlow
  • Hugging Face Transformers
  • LangChain
  • LlamaIndex
  • RAG
  • LoRA / QLoRA
  • Prompt engineering
  • OpenAI API
  • Anthropic Claude
  • AWS Bedrock
  • BERT
  • XGBoost
  • spaCy
  • Rasa
02

Platforms

  • Python
  • FastAPI
  • Flask
  • REST APIs
  • Microservices
  • Java
  • TypeScript
  • Kafka
  • Redis
03

Operations

  • Docker
  • Kubernetes
  • AWS EC2 / S3 / Lambda
  • SageMaker
  • GCP
  • Azure ML
  • MLflow
  • Weights & Biases
  • Prometheus
  • Grafana
  • Terraform
  • CI/CD
04

Data & retrieval

  • SQL
  • PostgreSQL
  • Spark
  • Pandas
  • NumPy
  • Pinecone
  • FAISS
  • Weaviate
04EXPERIENCE

Where I’ve made
the work count.

MAY 2024 - PRESENT

Netflix

AI / LLM Engineer

Building RAG recommendations, shipping QLoRA model adaptation, and operating low-latency ML inference services. Mentored 3 junior engineers on fine-tuning and RAG.

  • 10M+ daily requests
  • 99.9% uptime
  • <200ms response
MAY 2021 - JUL 2022

Cognizant

Software Engineer · AI/ML

Delivered NLP fraud detection, conversational support, document processing, and credit-risk ML pipelines. Built MLflow drift monitoring and reusable Spark feature pipelines.

  • 40% less fraud
  • 85% resolution rate
  • 60% less manual review
EDUCATION

M.S. Computer Science · Machine Learning & AI
Missouri University of Science and Technology · 2024

B.E. Electronics & Communication
Sathyabama University · 2021

05CONTACT

Have a hard
problem? Good.

I’m open to conversations about AI/ML engineering, applied LLM systems, and product-minded technical roles.

sairambodapothula2@gmail.com ↗

All fields are required. Your message is sent to my private inbox.