cs.AI · cs.SE · econ.GN · q-fin.GN · preprint · 20 May 2026

Engineered Intelligence:
A note on building above
what already works.

SS Sigma Infotech Pvt. Ltd.

Registered: Visakhapatnam, Andhra Pradesh. Studio: Hyderabad, Telangana. India.

CIN U62013AP2023PTC111323 · DPIIT DIPP226093 (valid 2025–2033) · MSME UDYAM-AP-10-0042250

Submitted 12 May 2026. Revised 20 May 2026. Open access. Correspondence — chaitanya@sssigma.in.

★ Read it without the Greek and Latin — brochure edition →

Abstract.

SS Sigma is a deep-tech IT services firm — incorporated June 2023, registered in Visakhapatnam, working out of Hyderabad — building agentic AI products and regulated-industry platforms for governments, institutions, and enterprises. We argue, with reference to six practices (§2) and the firm's working roadmap (§6), that the most economically rational way to introduce intelligence into a large institution is not to replace its working systems but to compose an intelligent layer above them (§3). This note is the firm's working position; §5 reports the verifiable founder-level track record on which the present work is built.

Keywords. agentic ai; intelligent layer; graph-RAG; post-quantum cryptography; edtech operating systems; skills marketplace; aviation operations; rbi-licensed p2p; india; deep tech.

§1. Introduction

This note is the working position of SS Sigma Infotech Pvt. Ltd. — a deep-tech IT services firm, incorporated June 2023 (registered office Visakhapatnam, founder's studio Hyderabad), that builds agentic AI products and regulated-industry platforms for governments, institutions, and enterprises.¹

We hold that the prevailing instinct in institutional IT — to replace legacy systems with newer, fully-integrated successors — is most often a mistake. Legacy systems are frequently the only honest historical record an institution still possesses; replacing them severs that record, costs years, and rarely delivers the intelligence the replacement promised. Our thesis, developed across the six practices enumerated in §2, is that intelligence should be composed as a layer — engineered above what already works, not in place of it (§3).

The remainder of this note is organised as follows. §2 introduces the firm's six practices and the kinds of problems each addresses. §3 states the intelligent-layer thesis formally and sketches three principles that follow from it. §4 summarises capabilities. §5 reports three pilot results from the period 2024–26. §6 gives correspondence. A short bibliography follows.

§2. Practices

The firm operates six practices. Each is a first-class product engagement, not a service line item; each ships its own software artefacts and operates its own pilot cycle. The set of practices $\mathscr{P}$ is enumerated below; the cardinality is:

$$\mathscr{P} \;=\; \bigl\{\, \mathrm{Shiksha\,Setu\,AI},\ \mathrm{Skill\,OS},\ \mathrm{Kavach\,Q},\ \mathrm{Aviation\,AI},\ \mathrm{FinTech},\ \mathrm{Enterprise\,IT} \,\bigr\}, \quad |\mathscr{P}| \;=\; 6.$$ (2.1)

§2.1. Shiksha Setu AI product · in build

The firm's flagship platform. An end-to-end agentic operating layer for NAAC-compliant colleges, university systems, and state-level government education programmes. Six modules — LMS, Virtual Labs, 3D Scholar, Teacher AI, Library, ERP — composed on a shared knowledge graph and identity layer. The teaching-assistant module (Teacher AI) is the platform's lead surface and the most-asked-for in early conversations.

§2.2. Skill OS product · in build

Adjacent to Shiksha Setu AI but oriented at a different surface: a skills-as-OS layer for institutional and government skill-development programmes. Skill OS provides assessment, credentialing, employer-side discovery, and a learner-portfolio identity that is portable across institutions. Designed to plug into existing ERP and learner-records systems rather than replace them — in line with §3 below.

§2.3. Kavach Q platform · in progress

Post-quantum cryptography for regulated BFSI and government. Kavach Q produces a Cryptographic Bill of Materials (CBOM) across a customer's stack, followed by a phased migration plan aligned to NIST PQC selections.² The dominant threat we model is store-now-decrypt-later; for institutions handling long-lived sensitive data, PQC migration is a deadline problem, not an option. An MoU with the AP Amaravati Quantum Mission is in negotiation.

§2.4. Aviation AI practice · in progress

Agentic intelligence for ground-side aviation operations: turnaround monitoring, crew-documentation workflows, predictive-maintenance signals, and ops analytics. The firm's aviation engagement begins on the operations side and is read-only on safety-critical buses — a constraint we treat as load-bearing. Pursued in coordination with the co-promoter's aerospace investment vehicle (Vaimanik Aerospace); SS Sigma supplies the agentic layer.

§2.5. FinTech practice · founder-led

Regulated payments and lending infrastructure. P2P escrow, KYC and CKYC, eNACH mandates, UPI handles, and connectors to core-banking systems via standard APIs. The practice carries the founder's RBI-NBFC-P2P licensing experience at FincSquare / FlexiP2P (one of only 26 RBI-licensed NBFC-P2Ps as of March 2024) and is the firm's source of operating discipline around audit trails and regulator interaction.³

§2.6. Enterprise IT

Cloud (AWS, Azure, GCP, sovereign), DevOps, QA automation, observability, and managed services. The practice exists to run what the other five build — and what the customer brought to the engagement — at a service level the customer is willing to underwrite. The founder's background in CMMI Level 5 QA and UAT automation for Fortune 500 retail (2013–17) underwrites the QA discipline.

Σ — summation operator. composition. layer. firm.

§3. The Intelligent-Layer Thesis

Institutional software, in our experience, fails through replacement more often than through age. A working system is a kind of evidence — of how the institution thinks, what it tolerates, where its accountability sits. Replace it, and you destroy that evidence faster than the successor can reproduce it. The intelligent-layer thesis is the alternative: leave the system, augment its inputs, intercept its outputs, and reason in between.

Proposition 3.1 (Intelligent-layer thesis).

Let $S$ denote a working institutional system. Let $\mathcal{A}$ denote an agent that observes the inputs to $S$, may read selected outputs of $S$, and may act only by writing back to $S$ through $S$'s existing interfaces. Let $C(t)$ denote the context available at time $t$. Then for a broad class of institutional tasks the composition

$$\sigma(t) \;=\; S\bigl(\mathrm{input}(t)\bigr) \;+\; \alpha \cdot \mathcal{A}\bigl(C(t)\bigr) \qquad \alpha \in [0,1] \tag{3.1}$$

dominates the strategy of replacing $S$ on three dimensions: (i) time to first measurable outcome, (ii) cumulative cost across a five-year horizon, and (iii) preserved institutional knowledge. The parameter $\alpha$ encodes the trust the institution is willing to place in $\mathcal{A}$; the firm's job is to make $\alpha$ auditable, not to make it large.

§3.1. Three Principles

Principle I — No rip-and-replace. Legacy portals, departmental databases and call centres stay alive. We connect through secure APIs and event buses, not through migrations no-one asked for. Where the legacy is the truth, we treat it as the truth.

Principle II — Pilot, then prove. Start with one ministry, one school district, one route, one workflow. Validate measurable outcomes against a real baseline — never against a hypothetical one — before any phased scale-up. If a pilot's effect is not measurable, the pilot has not run yet.

Principle III — Humans stay in charge. Agents assist intake, routing, summaries, and alerts. Officers, teachers, controllers, and bankers keep sensitive decisions. Authority is never the model's. We log every assist, every override, and every refusal, and we hand the log to the customer.

§4. Capabilities

Table 1 enumerates capabilities by phase of engagement (rows) and competency column (Strategy, Engineering, AI & Data, Operations). The grid is a working manifest; an engagement composes across phases and columns simultaneously rather than sequentially.

Table 1. Capabilities, indexed by phase of engagement and competency column.

Phase	Strategy	Engineering	AI & Data	Operations
αPlan	Digital-transformation strategy. Enterprise & solution architecture.	Application-modernisation discovery. Reference architectures.	Agent design, evaluation harnesses, ontology drafting.	Programme governance. Vendor & risk mapping.
βBuild	Government workshops & proposal engineering.	Cloud (AWS / Azure / GCP / sovereign). API gateways. Secure connectors.	Knowledge graph & GraphRAG. Voice AI. Document intelligence.	DevOps. CI/CD. Infra automation. Observability.
γRun	Advisory retainer. Solution evolution.	QA automation. Performance & resilience engineering.	Personalised models. Vector & analytics stores.	SmartOps (AIOps). Managed services. 24×7 SRE.

§5. Track Record

SS Sigma was incorporated in June 2023; the firm's first commercial year (FY 2024–25) closed at ₹18.33 lakhs of audited revenue.⁴ The bulk of operational evidence on which the firm's practices stand therefore sits with the founders' prior engagements, summarised below. Customer-attributable pilot outcomes from SS Sigma's own engagements will be reported in subsequent editions of this note.

Record 5.1 (Founder — RBI-licensed NBFC-P2P, FincSquare / FlexiP2P).

The founder led product at FincSquare, one of 26 RBI-licensed NBFC-P2P lenders in India as of March 2024, and was on the founding team that obtained the NBFC-P2P licence (late 2018). As Head of Product (2024–present), the engagement returned a 30% lift in loan originations and a 60% reduction in manual-verification time, both measured against a 90-day pre-pilot baseline. Escrow architecture, KYC / CKYC flows, and RBI-compliance plumbing were all engineered to audit-first discipline; this is the operating method the firm now ships into other regulated practices.

Record 5.2 (Founder — Unschool, EdTech, 2023–25).

Co-founder, Unschool (Apr 2023 – Nov 2025). Led B2B institutional offerings and the skills-marketplace surface. The engineering substrate of Shiksha Setu AI and Skill OS, particularly around institutional onboarding and credential issuance, is the direct technical descendant of that work.

Record 5.3 (Co-promoter — 34-year career in ONGC-tier industrial contracting).

The co-promoter operated Sreeveera Lakshmi Engineering & Fabricators as an ONGC A1 contractor for 34 years (1989–2023), with a portfolio across bunk houses, mud-tank systems, AMCs, and oil-field equipment services. The discipline that this lineage brings to SS Sigma is governance, cash-flow, and delivery assurance against contract — the unglamorous half of how regulated work actually ships. The co-promoter additionally chairs Vaimanik Aerospace, the firm's sister vehicle for aerospace and defence engagements.

§6. What We Are Building

Three products and two emerging practices, summarised by status. Each line resolves to a separate technical note as it matures; this section is intentionally a register, not a brochure.

6.1 Shiksha Setu AI (product, in build).

End-to-end agentic operating layer for NAAC-compliant colleges, university systems, and state-government education programmes. Six modules — LMS, Virtual Labs, 3D Scholar, Teacher AI, Library, ERP — composed on a shared knowledge graph. Target deployment surface: institutional pilot, FY26 H2.

6.2 Skill OS (product, in build).

Skills-as-OS layer: assessment, credentialing, employer-side discovery, and a portable learner-portfolio identity. Plugs into existing ERP and learner-records systems rather than replacing them. Target deployment: skills-mission and institutional partnerships, FY26 H2.

6.3 Kavach Q (platform, in progress).

Post-quantum cryptography platform for regulated BFSI and government. CBOM generation, phased migration planning aligned to NIST PQC selections, sandbox swap of TLS suites. An MoU with the AP Amaravati Quantum Mission is in negotiation; commercial structuring led by the co-promoter.

6.4 Aviation AI (practice, in progress).

Ground-side ops intelligence — turnaround monitoring, crew-documentation workflows, predictive-maintenance signals. Pursued in coordination with Vaimanik Aerospace; SS Sigma supplies the agentic layer. Read-only on safety-critical buses; writes confined to ops dashboards and document stores.

6.5 FinTech engineering & advisory (practice, founder-led).

RBI-licensed P2P infrastructure, escrow, KYC / CKYC, eNACH, UPI, and core-banking connectors. Sourced directly from the founder's tenure at FincSquare / FlexiP2P (Record 5.1); ongoing engagements run from the firm's Hyderabad studio.

6.6 Enterprise IT & managed services (practice, founder-led).

Cloud, DevOps, QA automation, observability, and 24×7 SRE. The firm's run-time backbone for everything in §6.1–6.5 above, and for legacy systems the customer brought.

§7. Correspondence

The firm welcomes correspondence from prospective customers (governments, institutions, banks), prospective colleagues (engineers, designers, researchers), and prospective collaborators (universities, regulators, peer firms). Write to chaitanya@sssigma.in — the founder will reply within two business days.

Founders. Chaitanya Vihari Uppalapati — Managing Director & Promoter. Chinteswara Rao Chennamsetti — Director & Co-promoter (finance, governance, aerospace).

Registered office. SS Sigma Infotech Pvt. Ltd., DN 39-21-40/1/151, 5F 512, Lansum Greens, Madhavadhara, Muralinagar, Visakhapatnam, Andhra Pradesh – 530007, India.

Studio. Hyderabad, Telangana. Visits by appointment.

References & Statutory Identity

National Institute of Standards and Technology (2024). Module-Lattice-Based Key-Encapsulation Mechanism Standard. FIPS 203. U.S. Department of Commerce, Gaithersburg MD.
National Institute of Standards and Technology (2024). Module-Lattice-Based Digital Signature Standard. FIPS 204. U.S. Department of Commerce.
National Institute of Standards and Technology (2024). Stateless Hash-Based Digital Signature Standard. FIPS 205. U.S. Department of Commerce.
Reserve Bank of India (2017). Master Directions — Non-Banking Financial Company – Peer-to-Peer Lending Platform (Reserve Bank) Directions, 2017. RBI/DNBR/2017-18/57.
Ministry of Education, Government of India (2020). National Education Policy 2020. Curricular and pedagogical reform across school and higher-education stages.
Department for Promotion of Industry and Internal Trade (2025). DPIIT Recognition Certificate — SS Sigma Infotech Pvt. Ltd., DIPP226093, valid 17 Oct 2025 – 01 Jun 2033. Government of India.
Ministry of Micro, Small & Medium Enterprises (2023). Udyam Registration — UDYAM-AP-10-0042250 (Micro, Services). Government of India.
Ministry of Corporate Affairs (2023). Certificate of Incorporation — SS Sigma Infotech Private Limited, CIN U62013AP2023PTC111323. ROC Vijayawada.

Acknowledgements. The firm thanks Dr Ravi Daryavesh and D Ravi & Co (ICAI FRN 025178S) for FY24–25 audit and statutory diligence; ICICI Bank, Visakhapatnam Main, for the firm's primary banking; and the engineers, designers, and partners whose work is the substance of every line above.